System

Abstract

A system is provided.SOLUTION: A system including means for collecting purchase information of customers, stock data, disposal data, and employee shift information, means for pre-processing the collected data and complementing missing values and abnormal values, means for performing demand prediction, optimization of inventory order quantity, optimization of price setting, and automatic generation of employee shifts using an AI model based on the pre-processed data, means for presenting necessary information to a user based on a prediction result or generated data and performing confirmation and correction, and means for automatically executing necessary actions based on the information confirmed and corrected by the user.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] Food waste, labor shortages, and low wage systems are major issues in the modern service industry. These issues reduce the efficiency of business operations and make it difficult to provide sustainable services. Excess inventory and mismatches between supply and demand can also lead to lower profit margins and increased waste. Even in employee shift management, manual scheduling is time-consuming and labor-intensive, making it difficult to allocate appropriate personnel. There is a need to solve these issues and achieve efficient and sustainable business operations. [Means for solving the problem]

[0005] The present invention provides a means for collecting customer purchasing information, inventory data, waste data, and employee shift information and preprocessing this data. It also provides a system that uses an AI model to forecast demand, optimize inventory order quantities, optimize pricing, and automatically generate employee shifts based on the collected and preprocessed data. This system presents necessary information to the user based on the prediction results and generated data, allowing them to confirm and modify it. Necessary actions are then automatically taken based on the information confirmed and modified by the user. In this way, food waste reduction, mitigation of labor shortages, and appropriate pricing become possible, resulting in the provision of sustainable services.

[0006] "Customer purchasing information" refers to data about products and services that a specific customer has purchased in the past.

[0007] "Inventory Data" refers to information regarding the quantity of goods or materials held at a particular time.

[0008] "Waste data" refers to information about the types and quantities of products that are discarded without being used.

[0009] "Employee shift information" refers to data indicating when each employee is scheduled to work during a particular time period.

[0010] "Preprocessing" refers to the process of formatting the collected raw data and completing or removing missing or outlier values.

[0011] An "AI model" refers to an algorithm or calculation method that uses artificial intelligence technology to analyze data and make predictions and optimizations.

[0012] "Demand forecasting" refers to the process of predicting future demand based on past data.

[0013] "Inventory order quantity optimization" refers to calculating appropriate order quantities based on forecasted demand to prevent excess inventory or shortages.

[0014] "Pricing optimization" refers to the process of determining the appropriate price for a product or service by taking into account the balance between supply and demand.

[0015] "Automatic employee shift generation" refers to the automatic creation of employee shift schedules using AI models. [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0017] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0018] First, the terms used in the following description will be explained.

[0019] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0020] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0021] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0022] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0023] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0024] [First embodiment]

[0025] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0026] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0027] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0028] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0029] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0030] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0031] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0032] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0034] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0035] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0036] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0037] This invention relates to a system for optimizing specific business operations, and aims to improve business efficiency by applying AI models to mainly utilize customer purchasing information, inventory data, waste data, and employee shift information. The system consists of three main components: a server, a terminal, and a user.

[0038] Automated ordering system

[0039] Server: In the ordering system, the server first collects inventory status, past order data, and waste data. Next, the server preprocesses this data and fills in missing and outlier values. Based on the preprocessed data, the server uses an AI model to forecast demand and calculate the next order quantity.

[0040] For example, if an analysis of data from the past year predicts that demand for a particular product will increase next week, the server automatically generates an appropriate order quantity based on that prediction. After the user checks this order list and makes any necessary corrections, the server automatically sends the final list to the supplier.

[0041] Dynamic Pricing System

[0042] Server: In this system, the server collects customer purchase history and store visits by time of day, and pre-processes this data. Based on the pre-processed data, the server uses an AI model to analyze the relationship between demand and price. Next, it sets the optimal price based on demand and automatically distributes it to the POS system.

[0043] For example, if past purchase history shows that demand is highest between 3:00 PM and 6:00 PM on weekdays, you can set the price to be 10% higher during that time period, which will reduce unsold items and maximize profits.

[0044] Automatic shift creation

[0045] Server: In the shift creation system, the server collects employee shift history, desired vacation time, and store visit predictions. It then preprocesses this data and uses an AI model to predict optimal shifts based on the number of employees and peak hours. The predicted shift schedule is automatically generated and can be reviewed and modified by the user.

[0046] For example, if a high number of customers is predicted on a particular weekend, the system generates a shift schedule that automatically allocates the necessary number of employees during that time period. The user can check the generated schedule, correct any problems, and automatically distribute it to employees.

[0047] Service and product demand forecasting

[0048] Server: In a demand forecasting system, the server collects and preprocesses customer purchase history and sales data. Based on the preprocessed data, the server uses an AI model to predict future demand. Based on the prediction results, it generates a list of recommended promotional items and inventory replenishment and presents it to the user.

[0049] For example, if past data predicts that a particular product will be popular in the next month, the product will be listed as a promotion target. Users can check the list and make appropriate promotion plans and stock replenishment.

[0050] In this way, the invention will reduce food waste, alleviate labor shortages, and achieve sustainable service provision through appropriate pricing.

[0051] The processing flow will be explained below.

[0052] Automated ordering system processing

[0053] Step 1:

[0054] Server: Collects inventory status, past order data, and disposal data from the database.

[0055] Every day at 2:00 AM, the server collects inventory status, order data, and disposal data from the database for the past year.

[0056] Step 2:

[0057] Server: Preprocesses the collected data and imputes missing or outlier values.

[0058] After collecting the data, the server performs preprocessing, imputing missing values ​​with the average value and removing outliers.

[0059] Step 3:

[0060] Server: Preprocessed data is input into the AI ​​model to generate demand forecasts.

[0061] The server passes the pre-processed data to an AI model to generate demand forecasts for the next week.

[0062] Step 4:

[0063] Server: Calculates the optimal order quantity based on the demand forecast results.

[0064] The server calculates the optimal order quantity for each product based on the demand forecast results output by the AI ​​model.

[0065] Step 5:

[0066] Server: Automatically generate an order list based on the optimal order quantity.

[0067] The server creates an order list based on the calculated order quantity.

[0068] Step 6:

[0069] User: Check the generated order list and modify it as necessary.

[0070] The user checks the order list on the system and corrects the quantity if necessary.

[0071] Step 7:

[0072] Server: Automatically sends the confirmed and corrected order list to the supplier.

[0073] The server receives the order list that the user has confirmed and modified, and automatically sends it to the specified supplier.

[0074] Dynamic pricing system processing

[0075] Step 1:

[0076] Server: Collects customer purchase history and store visit data by time of day.

[0077] Every night, the server collects customer purchase history and time-of-day visit data from the database.

[0078] Step 2:

[0079] Server: Preprocesses the collected data and removes outliers.

[0080] The server performs preprocessing after data collection to normalize or remove outliers.

[0081] Step 3:

[0082] Server: Inputs the preprocessed data into the AI ​​model and analyzes the relationship between demand and price.

[0083] The server passes the pre-processed data to an AI model that analyzes the relationship between demand and price.

[0084] Step 4:

[0085] Server: Sets the price for the next day based on the analysis results.

[0086] The server uses the analysis results output by the AI ​​model to set the optimal price for the next day.

[0087] Step 5:

[0088] Server: Automatically distributes the set price information to the POS system.

[0089] The server sends the set price information to the POS system, which updates it automatically.

[0090] Process automatic shift creation

[0091] Step 1:

[0092] Server: Collects employee shift history, desired vacation time, and store visit forecast data.

[0093] At the end of each month, the server collects employee shift history, desired vacation data, and store visit forecast data from the database.

[0094] Step 2:

[0095] Server: Preprocesses the collected data and imputes missing or outlier values.

[0096] After data collection, the server performs preprocessing, imputing missing values ​​and removing outliers.

[0097] Step 3:

[0098] Server: Based on the pre-processed data, the AI ​​model predicts the optimal shift.

[0099] The server inputs the pre-processed data into an AI model to predict the optimal shift schedule.

[0100] Step 4:

[0101] Server: Automatically generates predicted shift schedules.

[0102] The server automatically generates shift schedules based on the prediction results output by the AI ​​model.

[0103] Step 5:

[0104] User: Review the generated shift schedule and modify it if necessary.

[0105] Users can check the shift schedule on the system and modify it if necessary.

[0106] Step 6:

[0107] Server: Automatically distributes confirmed shift schedules to employees.

[0108] The server receives the shift schedule that the user has confirmed and modified and automatically distributes it to all employees.

[0109] Processing demand forecasts for services and products

[0110] Step 1:

[0111] Server: Collects customer purchase history and sales data.

[0112] The server collects customer purchase history and sales data from the database every day at 11:00 p.m.

[0113] Step 2:

[0114] Server: Preprocesses the collected data and imputes missing or outlier values.

[0115] After data collection, the server performs preprocessing, imputing missing values ​​and removing outliers.

[0116] Step 3:

[0117] Server: Inputs the pre-processed data into the AI ​​model to predict future demand.

[0118] The server passes the pre-processed data to an AI model to predict future demand.

[0119] Step 4:

[0120] Server: Based on the prediction results, a list of recommended promotion targets and inventory replenishment is generated.

[0121] The server automatically generates a list of recommended promotional targets and inventory replenishment based on the prediction results output by the AI ​​model.

[0122] Step 5:

[0123] User: Review the generated recommendation list and incorporate it into promotion planning and inventory replenishment.

[0124] The user checks the recommended list and implements promotion plans and inventory replenishment as necessary.

[0125] Example 1

[0126] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0127] In conventional business systems, customer purchasing information, inventory data, disposal data, and employee shift information are often managed separately, making it difficult to integrate and efficiently utilize this data. Furthermore, business processes such as demand forecasting, pricing, and automatic shift generation are performed manually, requiring a great deal of time and effort, making it difficult to improve the efficiency and accuracy of business operations. Aiming to solve these problems, the present invention provides a system for optimizing specific business operations.

[0128] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0129] In this invention, the server includes means for collecting customer purchasing information, inventory data, disposal data, and employee shift information, means for preprocessing the collected data and complementing missing values ​​and outliers, means for using a generative AI model based on the preprocessed data to perform demand forecasting, optimizing inventory order quantities, optimizing pricing, and automatically generating employee shifts, means for presenting necessary information to a user based on the prediction results and generated data and for confirmation and correction, and means for automatically executing necessary actions based on the information confirmed and corrected by the user. This enables efficient management of customer purchasing information, inventory data, disposal data, and employee shift information, automating business processes such as demand forecasting, inventory ordering, pricing, and shift creation, and improving business efficiency and accuracy.

[0130] Key Word Definitions

[0131] "Customer purchasing information" refers to data that records the type, quantity, purchase date, etc. of products that a customer has purchased in the past.

[0132] "Inventory data" refers to data for managing the quantity of products currently in stock, storage locations, and inventory history.

[0133] "Disposal data" refers to data that records the quantity of products that have been discarded due to reasons such as expiration or poor quality, the reason for disposal, and the date and time of disposal.

[0134] "Employee shift information" refers to data necessary for labor management, such as employees' working hours, desired holidays, and shift history.

[0135] "Missing values" refers to a state in which data that should be present in a dataset is missing.

[0136] An "outlier" is a value that is significantly different from other data in a data set and is typically treated as an error or outlier.

[0137] A "generative AI model" is a model that uses artificial intelligence technology to learn specific patterns from data and automatically perform tasks such as prediction and classification.

[0138] "Demand forecasting" is the process of predicting future demand based on past data, and refers to predicting sales quantities, usage amounts, etc.

[0139] "Inventory order optimization" is the process of determining optimal inventory levels based on demand forecasts to avoid shortages and excess inventory.

[0140] "Pricing optimization" is the process of setting optimal selling prices taking into account demand and market conditions, with the goal of maximizing profits.

[0141] "Automatic employee shift generation" is the process of automatically creating employee work schedules, taking into account business demands and employee preferences.

[0142] "Means for automatically taking necessary actions based on information confirmed and corrected by the user" refers to a process for automatically taking business actions such as placing orders, changing prices, and notifying shifts based on data confirmed and corrected by the user.

[0143] MODE FOR CARRYING OUT THE INVENTION

[0144] This invention relates to a system for optimizing specific business operations, and aims to improve business efficiency by applying generative AI models primarily using customer purchasing information, inventory data, waste data, and employee shift information. This system consists of three main components: a server, a terminal, and a user.

[0145] Hardware and software used

[0146] The server is built on AWS (a common cloud computing service) and uses MySQL as the database. Python and libraries such as Pandas and Scikit-learn are used for data preprocessing and analysis. AI frameworks such as TensorFlow, Keras, XGBoost, and Prophet are mainly used to implement the generative AI model. A general sales management system is assumed as the POS system.

[0147] System example and operation

[0148] Automated ordering system

[0149] Server: The server periodically collects inventory status, past order data, and waste data from terminals and point-of-sale (POS) systems, and preprocesses this data using Python and Pandas. It imputes missing and outlier values ​​and inputs the data into a generative AI model (e.g., TensorFlow) to perform demand forecasting. It automatically generates an order list based on the forecast results and presents it to the user. The user can review and modify the order list, and the final list is automatically sent from the server to the supplier.

[0150] Example: Based on data from the past year, the order quantity for the next week, when demand is predicted to increase, is automatically calculated and an order list is created.

[0151] Dynamic Pricing System

[0152] Server: The server collects and preprocesses customer purchase history and store visit status by time of day in real time. This data is input into a generative AI model (e.g., XGBoost) to analyze the relationship between demand and price. The optimal price is calculated from the analysis results and automatically distributed to the sales management system.

[0153] Example: If analysis shows that demand is high between 3pm and 6pm on weekdays, the price during that time period will be increased by 10%.

[0154] Automatic shift creation

[0155] Server: The server collects and preprocesses employee shift history, desired vacation time, and store visit prediction information. This data is input into a generative AI model (e.g., Keras) to predict and automatically generate employee shift schedules. Users can check and modify the generated shift schedules. The final shift schedules are automatically distributed from the server to employees.

[0156] Example: If store visits are predicted to be high on a particular weekend, create a shift that places more employees at that time.

[0157] Service and product demand forecasting

[0158] Server: The server periodically collects customer purchase history and sales data, and after preprocessing, inputs it into a generative AI model (e.g., Prophet) to predict future demand. Based on the prediction results, it generates a list of recommended promotional items and inventory replenishment and presents it to the user. The user checks this list and takes the necessary actions.

[0159] Example: If historical data predicts that a particular product will be popular next month, target that product for promotion.

[0160] Prompt Sentence Examples

[0161] "Please forecast demand for the next month based on data from the past year, and propose periods when demand will be high and product promotions. Please also let us know specific promotion ideas."

[0162] In this way, the present invention is a system that uses a generative AI model to integrate and analyze a variety of business data and automatically implement optimized business processes, thereby improving business efficiency and supporting highly accurate decision-making.

[0163] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0164] Processing steps of specific business operation optimization system

[0165] Automated ordering system processing flow

[0166] Step 1: Data collection

[0167] Server: The server periodically collects inventory status, past order data, and waste data from terminals and POS systems. Raw data obtained from each terminal is sent to the server as input. The output is consolidated data stored in a database within the server. As a concrete example, an automatic data collection script is executed every night.

[0168] Step 2: Data Preprocessing

[0169] Server: The collected data is preprocessed using Python and Pandas. The collected raw data is used as input. Missing values ​​and outliers are imputed, and the output is formatted as time series data. For example, missing values ​​are imputed with the past average value, and outliers with the median value.

[0170] Step 3: Demand forecasting using AI models

[0171] Server: The preprocessed data is input into a generative AI model (e.g., TensorFlow) to generate a demand forecast. The forecast model takes into account trends and seasonal fluctuations based on past data. The preprocessed data is used as input, and a forecast of future demand is output. For example, if demand for next week is predicted to be 200 units, the server outputs that number.

[0172] Step 4: Generate an order list

[0173] Server: Automatically generates an order list based on the forecast results. Based on the forecasted demand, the optimal order quantity is calculated taking into account the safety stock amount, and compiled in list format. The demand forecast results are used as input, and the output is an order list. As a specific example, if the forecasted demand is 200 units, an order list for 220 units is generated taking into account the safety stock amount.

[0174] Step 5: Check and modify the order list

[0175] User: The user checks the generated order list in a dedicated interface and modifies it if necessary. The automatically generated order list is used as input, and a modified final list is generated as output. For example, if the user feels that there are too many of a particular product in this list, the order quantity can be adjusted by the user.

[0176] Step 6: Send the order list to your suppliers

[0177] Server: Automatically sends the final order list, which has been reviewed and modified by the user, to the supplier. It uses the modified order list as input and generates the order instructions that are sent to the supplier as output, for example, via email or electronic data interchange (EDI).

[0178] Dynamic pricing system processing flow

[0179] Step 1: Data collection

[0180] Server: The server collects customer purchase history and store visit status by time of day in real time. The input is raw data obtained from the POS system, and the output is purchase data stored in a database within the server. For example, the latest data is obtained from the POS system every hour.

[0181] Step 2: Data Preprocessing

[0182] Server: Cleanses the collected data and removes outliers using Python and Scikit-learn. The input is raw data collected from the POS system, and the output is a clean dataset. For example, the format of dates and times is standardized.

[0183] Step 3: Analyze the relationship between demand and price

[0184] Server: The preprocessed data is input into an AI model (e.g., XGBoost) to analyze the relationship between demand and price. The input is the preprocessed data, and the output is the analysis result. For example, demand forecasting is performed using the relationship between past price fluctuations and demand.

[0185] Step 4: Set the optimal price

[0186] Server: Sets the optimal price based on the analysis results. The analysis results are used as input and new pricing is generated as output. For example, the price is set to increase by 10% during times of high demand.

[0187] Step 5: Distributing price information to POS systems

[0188] Server: Calculated price information is sent to the POS system in real time, automatically updating the store's prices. The input is the new price setting, and the output is the price information reflected in the POS system. For example, the new price information is immediately applied to the POS system and displayed to the customer.

[0189] Process flow for automatic shift creation

[0190] Step 1: Data collection

[0191] Server: The server collects employee shift history, desired vacation time, and store visit prediction information. The employee database and store visit prediction data are used as input, and the output is the integrated data stored in the server. For example, the latest data is obtained every Sunday.

[0192] Step 2: Data Preprocessing

[0193] Server: Cleanses the collected data and processes missing and outliers using Python and Pandas. The input is raw data, and the output is a clean dataset. For example, missing values ​​are imputed with the mean and outliers are removed.

[0194] Step 3: Shift prediction using AI models

[0195] Server: Input the preprocessed data into an AI model (e.g., Keras) to predict employee shifts. The input is the preprocessed data, and the output is the predicted shift schedule. For example, it allocates the required number of employees according to peak hours.

[0196] Step 4: Generate a shift schedule

[0197] Server: Automatically generates shift schedules based on the prediction results. The input is the prediction results, and the output is the generated shift schedule. As a specific example, shifts are adjusted according to the predicted number of customers.

[0198] Step 5: Check and modify the shift schedule

[0199] User: The user checks the generated shift schedule and modifies it if necessary. The input is the automatically generated shift schedule, and the output is the final modified schedule. For example, if a user requests sudden leave, the shift can be covered by another employee.

[0200] Step 6: Distribute shift schedules to employees

[0201] Server: Automatically distributes the final shift schedule to employees. The input is the revised shift schedule, and the output is the shift information sent to employees. For example, it can be sent via email or a dedicated app.

[0202] Service / product demand forecasting process flow

[0203] Step 1: Data collection

[0204] Server: The server periodically collects customer purchase history and sales data. The input is raw data obtained from the POS system, and the output is sales data stored in a database on the server. For example, data is collected at the end of each month.

[0205] Step 2: Data Preprocessing

[0206] Server: Cleanses the collected data and processes missing values ​​and outliers using Python and Pandas. The input is raw data, and the output is a clean dataset. For example, missing values ​​are imputed with the average value of the previous month, and outliers are removed.

[0207] Step 3: Demand forecasting using AI models

[0208] Server: Inputs the preprocessed data into a generative AI model (e.g., Prophet) to predict future demand. The input is the preprocessed data, and the output is the prediction result. For example, a specific product is expected to be in high demand next month.

[0209] Step 4: Generate a recommendation list

[0210] Server: Based on the prediction results, a list of recommendations for promotions and inventory replenishment is generated. The input is the prediction results, and the output is the generated recommendation list. As a concrete example, it lists products with increasing demand.

[0211] Step 5: Review and modify the recommendation list

[0212] User: The user reviews the generated recommendation list and modifies it if necessary. The input is the automatically generated recommendation list, and the output is the modified list. For example, the user determines that a critical item needs to be restocked.

[0213] Step 6: Planning promotions and replenishing inventory

[0214] User: After reviewing the recommendation list, create promotion plans or replenish inventory based on the list. The input is the revised recommendation list, and the output is the implemented promotions or replenished inventory. For example, target specific products for a special campaign.

[0215] (Application example 1)

[0216] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0217] In traditional brick-and-mortar stores, inventory management, pricing, demand forecasting, and employee shift planning are often performed manually, resulting in problems such as inefficiency and human error in each process. Furthermore, there was a lack of means to grasp trends in customer purchasing behavior in real time and respond quickly, making appropriate inventory management and pricing difficult. Furthermore, it was difficult to implement automatic ordering based on forecasts or dynamic pricing to prevent unsold items, making it difficult to develop effective sales strategies. It is essential to solve these problems and improve the overall efficiency of store operations.

[0218] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0219] In this invention, the server includes: means for collecting customer purchasing information, inventory data, disposal data, and employee shift information; means for preprocessing the collected data and complementing missing and outlier values; means for using an AI model based on the preprocessed data to forecast demand, optimize inventory order quantities, optimize pricing, and automatically generate employee shifts; means for presenting necessary information to a user based on the forecast results and generated data and allowing the user to confirm and correct it; means for automatically executing necessary actions based on the information confirmed and corrected by the user; means for providing inventory management functions in real time and issuing alerts to prevent stockouts and excess inventory; means for dynamically setting product prices based on purchase history and the number of customers by time period to prevent unsold items and maximize profits; means for automatically creating optimal shift schedules based on employee shift history and desired vacation days; and means for predicting future demand based on customer purchase history and sales data and generating recommended lists of promoted products and inventory replenishment. This allows for centralized management of various processes related to store operations, enabling efficient and accurate inventory management, pricing, demand forecasting, and shift creation.

[0220] "Customer purchasing information" is data including the type and quantity of products purchased by customers who visit a store, the date and time of purchase, and the attributes of the purchaser.

[0221] "Inventory data" refers to information relating to the current quantity of products held in the store, storage locations, replenishment status, and the like.

[0222] "Disposal data" is data indicating the type, quantity, and date and time of disposal of products that have been discarded due to the expiration of their sales period or deterioration of quality.

[0223] "Employee shift information" is data including an employee's working date and time, working hours, vacation days, desired work shifts, etc.

[0224] "Preprocessing" is the process of converting collected data into a format suitable for analysis and model training, and filling in missing or outlier values.

[0225] An "AI model" is a mathematical model that uses artificial intelligence technologies such as machine learning and deep learning to make useful predictions and classifications from data.

[0226] "Demand forecasting" is the process of predicting future demand for a product based on past data.

[0227] "Inventory order quantity optimization" is the process of preventing excess inventory and out-of-stock by predicting product demand and calculating appropriate order quantities based on that.

[0228] "Pricing optimization" is the process of adjusting the price of a product to an optimal value based on demand data, etc.

[0229] "Automatic generation of employee shifts" is the process of automatically creating optimal shift schedules based on employees' shift history, desired working hours, and store business demands.

[0230] "Prediction results and generated data" refers to information such as demand forecast results, optimal order quantities, appropriate prices, and shift schedules generated by AI models.

[0231] The "inventory management function" monitors inventory status in real time and reduces the risk of stockouts and excess inventory.

[0232] "Dynamic setting" means that values ​​and parameters can be flexibly changed according to real-time data and the current situation.

[0233] "Promotional Products" means selected products that are specially promoted or discounted for promotional purposes.

[0234] A "recommended list" is a list generated by the system to encourage specific actions or responses.

[0235] This invention is built as a system for optimizing specific business operations. The system consists of three main components: a server, a terminal, and a user.

[0236] Inventory management function

[0237] The server first collects customer purchasing information, inventory data, waste data, and employee shift information. This data is obtained from databases and POS systems. The server then preprocesses the data and fills in missing and outlier values. Specifically, it cleanses the data using data analysis libraries such as Pandas. Using the preprocessed data, it trains a demand forecasting model using AI libraries such as TensorFlow and Keras. This model predicts future inventory demand and calculates appropriate order quantities.

[0238] Dynamic Pricing Feature

[0239] The server also analyzes purchase history and store visits by time of day to determine optimal pricing based on demand. This data is also preprocessed and fed into the AI ​​model, which analyzes the relationship between demand and price and calculates the most profitable price. This price information is automatically sent to the POS system, where prices are adjusted in real time.

[0240] Automatic shift creation

[0241] The server automatically generates optimal shift schedules based on employees' shift history, desired holidays, and predicted store visits. Employee information and store visit prediction data are fed into an AI model to calculate the optimal shift schedule. Users can check and modify the generated schedule, and the final shift schedule is distributed to employees.

[0242] Demand forecasting and promotions

[0243] The server also collects customer purchase history and sales data to predict future demand. Based on this prediction, it targets specific products for promotion and generates a list of recommended inventory replenishment. Specifically, if the AI ​​model predicts that a particular product will be popular in the next month, it will list that product as a promotion target.

[0244] Hardware and software used

[0245] The hardware used is a server for database hosting and AI model training, and a smartphone as a client terminal. The software used is Pandas (data collection and preprocessing) and TensorFlow / Keras (demand forecasting AI model construction and training).

[0246] Examples of specific examples and prompts

[0247] For example, if a particular product is predicted to be popular in the coming month, the product may be listed for promotion and the store may be notified.

[0248] Example prompt sentence:

[0249] "Using past sales data, the AI ​​model predicts which products will be in high demand for the next month, and creates a promotion plan for those products."

[0250] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0251] Step 1:

[0252] The server collects customer purchasing information, inventory data, disposal data, and employee shift information.

[0253] These data are obtained from databases and POS systems. The collected data is first converted into a data frame using the Pandas library, and missing and outlier values ​​are imputed (data cleansing).

[0254] Input: Customer purchasing information, inventory data, disposal data, employee shift information

[0255] Output: Preprocessed data frame

[0256] Step 2:

[0257] The server trains the AI ​​model based on the preprocessed data.

[0258] Build and train a demand forecasting model using TensorFlow and Keras libraries. Historical purchase and inventory data are used as input data for the model to generate demand forecasts.

[0259] Input: Preprocessed data frame

[0260] Output: A trained demand forecasting model

[0261] Step 3:

[0262] The server uses the trained AI model to predict future inventory demand and calculate appropriate order quantities.

[0263] Past purchasing data is input into the forecasting model to predict demand for the next week or month. Based on the forecast results, the optimal inventory order quantity is calculated.

[0264] Inputs: Trained demand forecasting model, historical purchase data

[0265] Output: Future inventory demand forecast, order quantity

[0266] Step 4:

[0267] The server automatically generates an order list based on the demand forecast results and the order quantity, and presents it to the user.

[0268] The generated order list is displayed on the terminal for the user to review and modify, and after the user makes the appropriate modifications, the information is automatically sent to the supplier.

[0269] Input: Future inventory demand forecast, order quantity

[0270] Output: Order list, order data to suppliers

[0271] Step 5:

[0272] The server analyzes purchase history and store visits by time of day, and sets optimal prices based on demand.

[0273] The AI ​​model analyzes the relationship between demand and price to calculate the optimal price, which is then automatically sent to the POS system.

[0274] Input: Purchase history, store visits by time period

[0275] Output: Best price, price information to POS system

[0276] Step 6:

[0277] The server automatically generates an optimal shift schedule based on employee shift history, desired holidays, and predicted store visits, and presents it to the user.

[0278] The AI ​​model calculates the optimal shift schedule, which the user can then review and modify. The modified schedule is then distributed to employees.

[0279] Input: Shift history, desired leave, store visit forecast

[0280] Output: Shift schedule, information distributed to employees

[0281] Step 7:

[0282] The server predicts future demand based on customer purchase history and sales data, generates a recommended list of promotional items and inventory replenishment, and presents it to the user.

[0283] Based on the generated recommendation list, the user plans promotions and stock replenishment.

[0284] Input: Purchase history, sales data

[0285] Output: Demand forecast, recommendation list, promotional products

[0286] The above are the specific processing steps for carrying out the present invention.

[0287] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0288] This invention relates to a system for optimizing specific business operations, primarily using customer purchasing information, inventory data, waste data, and employee shift information to apply AI models to improve business efficiency. Furthermore, by combining it with an emotion engine that recognizes user emotions, even more accurate predictions and optimization can be achieved. This system consists of three main components: a server, a terminal, and a user.

[0289] Automated ordering system

[0290] Server: In the ordering system, the server first collects inventory status, past order data, and waste data. Next, the server preprocesses this data and fills in missing and outlier values. Based on the preprocessed data, the server uses an AI model to forecast demand and calculate the next order quantity. It also uses an emotion engine to incorporate user feedback and adjust the optimal order quantity.

[0291] For example, if an analysis of data from the past year predicts that demand for a particular product will increase next week, the system automatically generates an appropriate order quantity based on that prediction. After the user reviews this order list and the emotion engine analyzes the user's reactions and makes any necessary adjustments, the server automatically sends the final list to the supplier.

[0292] Dynamic Pricing System

[0293] Server: In this system, the server collects customer purchase history and store visits by time of day, and pre-processes this data. Based on the pre-processed data, the server uses an AI model to analyze the relationship between demand and price. Next, the server sets the optimal price based on demand, and an emotion engine incorporates user feedback to adjust the price setting, which is then automatically distributed to the POS system.

[0294] For example, if past purchasing history shows that demand is high between 3:00 PM and 6:00 PM on weekdays, the price can be increased by 10% during that time period, reducing unsold items and maximizing profits. The emotion engine analyzes user reactions and readjusts prices based on feedback.

[0295] Automatic shift creation

[0296] Server: In the shift creation system, the server collects employee shift history, desired vacation time, and store visit predictions. It then preprocesses this data and uses an AI model to predict optimal shifts based on the number of employees and peak hours. The predicted shift schedule is automatically generated and confirmed by the user. An emotion engine also analyzes the user's reactions and adjusts the shift schedule as necessary.

[0297] As a specific example, if a high number of customers is predicted on a particular weekend, the system will automatically generate a shift schedule that allocates the necessary number of employees during that time period. The user can check the generated schedule, and the emotion engine will analyze the user's reactions, correct any problems, and automatically distribute the schedule to employees.

[0298] Service and product demand forecasting

[0299] Server: In a demand forecasting system, the server collects and preprocesses customer purchase history and sales data. Based on the preprocessed data, the server uses an AI model to predict future demand. Based on the prediction results, it generates a list of recommended promotional items and inventory replenishment and presents it to the user. An emotion engine analyzes user reactions and adjusts the recommended list.

[0300] For example, if past data predicts that a particular product will be popular in the next month, the product will be listed as a promotion target. Users can check the list, and the emotion engine will analyze their reactions and make appropriate adjustments, such as planning promotions and replenishing inventory.

[0301] In this way, the invention will realize sustainable service provision through food waste reduction, labor shortage mitigation, and appropriate pricing. In addition, by integrating an emotion engine, more accurate prediction and optimization will be possible, leading to improved user satisfaction.

[0302] The processing flow will be explained below.

[0303] Automated ordering system processing

[0304] Step 1:

[0305] Server: Collects inventory status, past order data, and disposal data from the database.

[0306] Every day at 2:00 AM, the server collects inventory status, order data, and disposal data from the database for the past year.

[0307] Step 2:

[0308] Server: Preprocesses the collected data and imputes missing or outlier values.

[0309] After collecting the data, the server performs preprocessing, imputing missing values ​​with the average value and removing outliers.

[0310] Step 3:

[0311] Server: Preprocessed data is input into the AI ​​model to generate demand forecasts.

[0312] The server passes the pre-processed data to an AI model to generate demand forecasts for the next week.

[0313] Step 4:

[0314] Server: Calculates the optimal order quantity based on the demand forecast results.

[0315] The server calculates the optimal order quantity for each product based on the demand forecast results output by the AI ​​model.

[0316] Step 5:

[0317] Server: Automatically generate an order list based on the optimal order quantity.

[0318] The server creates an order list based on the calculated order quantity.

[0319] Step 6:

[0320] User: Check the generated order list and modify it as necessary.

[0321] The user checks the order list on the system and corrects the quantity if necessary.

[0322] Step 7:

[0323] Emotion Engine: Analyzes user reactions and suggests adjustments to the order list.

[0324] The emotion engine analyzes user reactions in real time and suggests adjustments to order quantities.

[0325] Step 8:

[0326] Server: Automatically sends the confirmed and corrected order list to the supplier.

[0327] The server receives the order list that the user has confirmed and modified, and automatically sends it to the specified supplier.

[0328] Dynamic pricing system processing

[0329] Step 1:

[0330] Server: Collects customer purchase history and store visit data by time of day.

[0331] Every night, the server collects customer purchase history and time-of-day visit data from the database.

[0332] Step 2:

[0333] Server: Preprocesses the collected data and removes outliers.

[0334] The server performs preprocessing after data collection to normalize or remove outliers.

[0335] Step 3:

[0336] Server: Inputs the preprocessed data into the AI ​​model and analyzes the relationship between demand and price.

[0337] The server passes the pre-processed data to an AI model that analyzes the relationship between demand and price.

[0338] Step 4:

[0339] Server: Sets the price for the next day based on the analysis results.

[0340] The server uses the analysis results output by the AI ​​model to set the optimal price for the next day.

[0341] Step 5:

[0342] Emotion Engine: Analyzes user reactions and suggests pricing adjustments.

[0343] The sentiment engine analyzes user reactions in real time and suggests pricing adjustments.

[0344] Step 6:

[0345] Server: Automatically distributes the set price information to the POS system.

[0346] The server sends the set price information to the POS system, which updates it automatically.

[0347] Process automatic shift creation

[0348] Step 1:

[0349] Server: Collects employee shift history, desired vacation time, and store visit forecast data.

[0350] At the end of each month, the server collects employee shift history, desired vacation data, and store visit forecast data from the database.

[0351] Step 2:

[0352] Server: Preprocesses the collected data and imputes missing or outlier values.

[0353] After data collection, the server performs preprocessing, imputing missing values ​​and removing outliers.

[0354] Step 3:

[0355] Server: Based on the pre-processed data, the AI ​​model predicts the optimal shift.

[0356] The server inputs the pre-processed data into an AI model to predict the optimal shift schedule.

[0357] Step 4:

[0358] Server: Automatically generates predicted shift schedules.

[0359] The server automatically generates shift schedules based on the prediction results output by the AI ​​model.

[0360] Step 5:

[0361] User: Review the generated shift schedule and modify it if necessary.

[0362] Users can check the shift schedule on the system and modify it if necessary.

[0363] Step 6:

[0364] Emotion engine: Analyzes user reactions and suggests shift schedule adjustments.

[0365] The emotion engine analyzes user reactions in real time and suggests adjustments to shift schedules.

[0366] Step 7:

[0367] Server: Automatically distributes confirmed shift schedules to employees.

[0368] The server receives the shift schedule that the user has confirmed and modified and automatically distributes it to all employees.

[0369] Processing demand forecasts for services and products

[0370] Step 1:

[0371] Server: Collects customer purchase history and sales data.

[0372] The server collects customer purchase history and sales data from the database every day at 11:00 p.m.

[0373] Step 2:

[0374] Server: Preprocesses the collected data and imputes missing or outlier values.

[0375] After data collection, the server performs preprocessing, imputing missing values ​​and removing outliers.

[0376] Step 3:

[0377] Server: Inputs the pre-processed data into the AI ​​model to predict future demand.

[0378] The server passes the pre-processed data to an AI model to predict future demand.

[0379] Step 4:

[0380] Server: Based on the prediction results, a list of recommended promotion targets and inventory replenishment is generated.

[0381] The server automatically generates a list of recommended promotional targets and inventory replenishment based on the prediction results output by the AI ​​model.

[0382] Step 5:

[0383] User: Review the generated recommendation list and incorporate it into promotion planning and inventory replenishment.

[0384] The user checks the recommended list and implements promotion plans and inventory replenishment as necessary.

[0385] Step 6:

[0386] Emotion engine: Analyzes user reactions and suggests adjustments to the recommendation list.

[0387] The emotion engine analyzes user reactions in real time and makes suggestions to adjust the recommendation list.

[0388] In this way, the invention will realize sustainable service provision through food waste reduction, labor shortage mitigation, and appropriate pricing. In addition, by integrating an emotion engine, more accurate prediction and optimization will be possible, leading to improved user satisfaction.

[0389] Example 2

[0390] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0391] In conventional business operation systems, customer purchasing information, inventory data, disposal data, and employee shift information are managed separately, making data integration and management cumbersome and making efficient operation difficult. Furthermore, predictions and optimization based on this data are often performed manually, limiting the efficiency of operations. Furthermore, because feedback that takes user emotions into account is not reflected, optimization results may not meet on-site needs. Therefore, a system that can solve these issues all at once is needed.

[0392] The identification processing by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting customer purchasing information, inventory data, disposal data, and employee shift information, means for preprocessing the collected data and complementing missing values ​​and outliers, means for performing demand forecasting using a generative AI model based on the preprocessed data, means for optimizing order quantities, optimizing pricing, and automatically generating employee shifts based on the generated demand forecast, means for obtaining user feedback using a sentiment analysis engine based on the prediction results and the generated data and adjusting necessary information, and means for automatically executing necessary actions based on information confirmed and corrected by the user. This enables integrated data management and efficient business operations, and realizes highly accurate predictions and optimization that reflect user feedback.

[0393] "Customer purchasing information" refers to data such as the products purchased by the customer, the purchase date and time, and the quantity purchased.

[0394] "Inventory data" refers to data including information such as product inventory quantity, date of receipt, and storage location.

[0395] "Disposal data" refers to data that includes information such as the quantity of inventory items to be discarded, the reason for disposal, and the date and time of disposal.

[0396] "Employee shift information" refers to data such as employees' working days, working hours, and desired holidays.

[0397] "Means of collection" refers to the methods and equipment used to obtain the necessary information from databases and sensors.

[0398] "Preprocessing means" refers to a method for completing missing data values, eliminating outliers, and converting the data into a format suitable for analysis.

[0399] A "generative AI model" refers to a model that includes algorithms for demand forecasting and optimization from data using techniques such as machine learning and deep learning.

[0400] "Means for demand forecasting" refers to methods and systems that use AI models to estimate future demand.

[0401] "Order quantity optimization" refers to the process of determining the optimal product order quantity based on demand forecasts.

[0402] "Pricing optimization" refers to the process of determining the most effective price by taking into account the balance between supply and demand.

[0403] "Automatic employee shift generation" refers to a system that automatically creates work schedules based on employee preferences and business needs.

[0404] An "emotion analysis engine" refers to a technology that analyzes emotions from user input, facial expressions, and voice, and reflects that feedback in the system.

[0405] "Means for adjusting required information" refers to a method or system that adjusts predictions or optimization results based on user feedback obtained using a sentiment analysis engine.

[0406] "Means for automatically executing necessary actions" refers to a method or system whereby the system automatically executes the relevant business process based on the information confirmed or corrected by the user.

[0407] This invention relates to a system for optimizing specific business operations, which utilizes customer purchasing information, inventory data, waste data, and employee shift information to apply a generative AI model to improve business efficiency. Furthermore, by combining it with a sentiment analysis engine, it achieves even more accurate prediction and optimization.

[0408] System configuration

[0409] This system consists of three main components: the server, the terminal, and the user. Each component will be explained in detail below.

[0410] Data collection and preprocessing

[0411] The server first collects customer purchasing information, inventory data, disposal data, and employee shift information. A relational database (e.g., MySQL or PostgreSQL) is used. After collecting the data, the server preprocesses the data and fills in missing and outlier values. Specifically, data cleaning is performed using the Python pandas library.

[0412] Demand forecasting and optimization

[0413] Based on the preprocessed data, the server uses a generative AI model (e.g., TensorFlow, PyTorch) to generate demand forecasts. Based on the generated demand forecasts, the server optimizes order quantities, optimizes pricing, and automatically generates employee shifts. For example, it uses a time series analysis model to forecast demand.

[0414] Sentiment analysis and feedback adjustment

[0415] The user checks the generated prediction and optimization results. Here, a sentiment analysis engine (e.g., OpenAI's GPT-4 model) analyzes the user's feedback and makes adjustments as necessary. Specifically, the user can check and correct the results using a dashboard provided in the web application.

[0416] Autorun and Notifications

[0417] The server automatically executes the necessary actions based on the information confirmed and corrected by the user. For example, if an order list is confirmed, the server automatically sends the order to the supplier using an API. It also distributes price information to the POS system.

[0418] Specific examples

[0419] A concrete example of how this works is shown below:

[0420] 1. Automated ordering system:

[0421] After analyzing data from the past year, it predicts that demand for a particular product will increase next week. Based on this prediction, an order list is generated and confirmed by the user. A sentiment analysis engine analyzes the user's reaction and adjusts the order quantity to the optimum. Finally, the server automatically sends the order list to the supplier.

[0422] 2. Dynamic Pricing System:

[0423] Based on past purchase history, it predicts high demand during certain times of the day and suggests pricing for those times. The user checks the pricing, and the sentiment analysis engine analyzes their reactions, readjusts the price, and automatically sends it to the POS system.

[0424] 3. Automatic shift creation:

[0425] If a high number of customers are predicted on certain weekends, a shift schedule is generated that increases the number of employees required during those times. Users can check the schedule, and the sentiment analysis engine analyzes reactions, adjusts it as necessary, and distributes it to employees.

[0426] 4. Service and product demand forecast:

[0427] Based on past data, certain products are predicted to be popular in the next month and are listed as promotional items. Users check the list, and a sentiment analysis engine analyzes reactions and makes adjustments, planning promotions, and replenishing inventory.

[0428] Example prompt sentence:

[0429] 1. "Based on the sales data from the past year, forecast next week's demand and calculate the optimal order quantity."

[0430] 2. "Based on past customer purchasing history, identify times of day when demand is high and suggest pricing for those times."

[0431] 3. "Generate optimal shift schedules based on employee shift history and store visit predictions."

[0432] 4. "Analyze sales data and create a recommended list of products to promote and replenish inventory for next month."

[0433] This completes the description of the embodiment of the invention. This system enables integrated data management and efficient business operations, and realizes highly accurate predictions and optimization that reflect user feedback.

[0434] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0435] Step 1: Data collection

[0436] The server retrieves customer purchasing information, inventory data, disposal data, and employee shift information from the database. SQL queries are used as input, and raw data extracted from the database is obtained as output. Specifically, the server executes periodically scheduled jobs to retrieve the required data.

[0437] Step 2: Data Preprocessing

[0438] The server preprocesses the acquired raw data. The input contains the raw data extracted in step 1, and the output is the preprocessed clean data. Specific operations include imputing missing values ​​(for example, median imputation) and correcting outliers using the pandas library. Outlier detection uses common statistical methods (for example, z-score).

[0439] Step 3: Demand forecast

[0440] The server uses a generative AI model (for example, a time series analysis model using TensorFlow) based on the preprocessed data to make a demand forecast. The input is the data preprocessed in step 2, and the output is the demand forecast result. Specifically, past purchase data is input into the model to predict future demand. In this process, the time series data is analyzed using a model such as LSTM.

[0441] Step 4: Optimize order quantity

[0442] The server optimizes the order quantity based on the demand forecast results. The input is the demand forecast results obtained in step 3, and the output is an optimized order list. Specifically, the demand forecast results are input into an optimization algorithm (e.g., linear programming) to calculate the order quantity.

[0443] Step 5: Optimize pricing

[0444] The server optimizes pricing based on the demand forecast results. The input is the demand forecast results obtained in step 3, and the output is an optimized price list. Specifically, it uses an AI model (e.g., regression analysis) to analyze the relationship between demand and price and sets prices.

[0445] Step 6: Automatic shift creation

[0446] The server automatically generates employee shifts based on the demand forecast results. The inputs are the demand forecast results obtained in step 3 and employee desired shift information, and the output is an optimal shift schedule. Specifically, the optimal shifts are generated using a shift generation algorithm (e.g., a Bayesian network).

[0447] Step 7: User confirmation and feedback

[0448] The user checks the generated order list, price list, and shift schedule. The input is the data provided by the server, and the output is the user's feedback. In concrete terms, the user checks the data on the web application, and the sentiment analysis engine analyzes the user's feedback. For example, the user may enter text feedback such as "This order quantity is too large."

[0449] Step 8: Analyze feedback and adjust

[0450] The server uses an emotion analysis engine to analyze user feedback and adjust prediction and optimization results as necessary. The input is user feedback, and the output is adjusted order lists, price lists, and shift schedules. Specifically, the emotion analysis engine analyzes the user's text and facial expressions and adjusts the data to reflect the results.

[0451] Step 9: Autorun and Notifications

[0452] The server automatically takes the necessary actions based on the adjusted information and notifies the relevant parties. The input is the data adjusted in step 8, and the output is the sent order list, updated price information, and distributed shift schedules. Specific operations include sending order information to suppliers using APIs, updating price information in the POS system, and distributing shift information to employees.

[0453] (Application example 2)

[0454] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0455] Conventional business operation systems have difficulty in efficiently optimizing demand forecasting, inventory management, pricing, and employee shift scheduling, and also have problems in that they do not respond to the burden and emotions felt by users. For these reasons, there is a need to efficiently operate business operations in physical stores and improve user satisfaction.

[0456] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0457] In this invention, the server includes means for collecting customer purchasing information, inventory data, disposal data, and employee shift information, means for preprocessing the collected data and complementing missing and outlier values, means for using an AI model based on the preprocessed data to perform demand forecasting, inventory order optimization, pricing optimization, and automatic employee shift generation, means for presenting necessary information to a user based on the prediction results and generated data and for confirmation and correction, means for automatically performing necessary actions based on the information confirmed and corrected by the user, means for collecting emotion data and adjusting information based on the user's emotion, and means for providing and managing information in real time via mobile terminals and wearable devices, thereby enabling smart, emotion-conscious business operations.

[0458] "Customer purchasing information" refers to data such as information about the products purchased by the customer, the date and time of purchase, and frequency of purchase.

[0459] "Inventory data" refers to data that indicates the current inventory status of products in stores and warehouses.

[0460] "Disposal data" refers to data relating to the disposal status of a product, the reason for disposal, and the date and time of disposal.

[0461] "Employee shift information" refers to information such as the work schedules and desired holidays of store employees.

[0462] "Preprocessing" is the process of complementing missing or outliers in collected data and preparing it in a form suitable for input into an AI model.

[0463] An "AI model" is an algorithm that uses machine learning and deep learning to predict demand and optimize data.

[0464] "Demand forecasting" is the process of predicting how much and how little a customer will purchase of a particular product or service.

[0465] "Optimizing inventory order quantities" means determining the appropriate order quantity, neither too much nor too little, based on demand forecasts.

[0466] "Pricing optimization" means setting the optimal selling price for a product, taking into account supply and demand conditions.

[0467] "Automatic generation of employee shifts" means automatically creating optimal shift schedules by taking into account store visitor forecasts and employees' desired holidays.

[0468] "Emotion data" is data based on the recognition of the user's reactions and emotions.

[0469] "Mobile terminals and wearable devices" refers to portable devices such as smartphones, smart glasses, and head-mounted displays.

[0470] "Means for providing and managing information in real time" refers to a method for obtaining data instantly, providing information to users, and efficiently managing business operations.

[0471] This invention provides a system for optimizing the business operations of a physical store based on customer purchase information, inventory data, disposal data, and employee shift information. The operation of the system is described in detail below.

[0472] System configuration

[0473] This system mainly consists of four main components: a server, a mobile terminal, a wearable device, and a user. The roles of each are as follows:

[0474] Server Roles

[0475] The server comprehensively manages customer purchasing information, inventory data, disposal data, and employee shift information. Specifically, it collects and preprocesses this data and applies AI models to perform the following tasks:

[0476] Data collection and preprocessing

[0477] The server automatically collects various data and performs preprocessing to fill in missing and outlier values. Specific software used is the Python pandas library and scikit-learn.

[0478] Demand Forecasting and Optimization

[0479] Based on the preprocessed data, an AI model (for example, a deep learning model using TensorFlow or Keras) is used to predict demand for the following week and optimize order quantities and pricing.

[0480] Emotion data capture

[0481] It collects user emotional data and adjusts prediction results based on this data. The emotion engine uses an NLP model to analyze emotions from user text and voice input.

[0482] Autorun

[0483] Based on the information confirmed and corrected by the user, necessary actions (such as sending an order list or updating a shift schedule) are automatically performed.

[0484] Mobile and wearable devices

[0485] Mobile devices (such as smartphones) and wearable devices (such as smart glasses and head-mounted displays) serve as tools to provide users with information in real time and assist in management.

[0486] Real-time information provision

[0487] Inventory information, sales status, employee shift information, and more are instantly presented to users via mobile devices or wearable devices.

[0488] User Interface

[0489] These devices allow for voice and touch input as user interfaces, allowing users to easily check and modify information.

[0490] User operations

[0491] Users can check the information provided in real time and adjust order lists, pricing, and shift schedules as needed. These adjustments are fed back to the emotion engine and reflected in optimization.

[0492] Component Examples

[0493] A concrete example is an inventory management assistant for a brick-and-mortar store, where store staff can use smart glasses to check inventory information in real time, view items that need to be ordered, and forecast demand for specific times of day. The assistant also uses an emotion engine to collect staff emotions and automatically adjust forecasts accordingly.

[0494] Prompt Sentence Examples

[0495] To manage your store's inventory in real time, please provide us with a dataset that includes the following items:

[0496] Product ID

[0497] Product name

[0498] Current Stock

[0499] Inventory quantity one week ago

[0500] Disposal Data

[0501] Past sales data

[0502] In this way, collecting emotional data and optimizing business operations will enable efficient business operations in physical stores.

[0503] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0504] Step 1:

[0505] The server collects customer purchasing information, inventory data, waste data, and employee shift information through APIs. To obtain this data, the server uses HTTP requests to retrieve the necessary information from databases and external services. The input data is in JSON or CSV format, and the output is the collected raw data.

[0506] Step 2:

[0507] The server preprocesses the collected raw data. Specifically, it converts it into a data frame using the pandas library and imputes missing or outlier values. It also normalizes and standardizes the data as needed. This results in data in a format suitable for input to the AI ​​model. The input is the collected raw data, and the output is the preprocessed data.

[0508] Step 3:

[0509] The server applies AI models to the preprocessed data to forecast demand and optimize inventory order quantities. Specifically, it loads trained models using TensorFlow and Keras and executes predictions. The input is the preprocessed data, and the output is the prediction results.

[0510] Step 4:

[0511] The server collects the user's emotional data. The emotional data is obtained by analyzing the user's text and voice input with an NLP model. Specifically, it uses Hugging Face's Transformers for emotion analysis. The input is the user's text and voice data, and the output is the analyzed emotional data.

[0512] Step 5:

[0513] The server adjusts the prediction results based on the analyzed emotional data. By incorporating emotional data into the prediction results, more realistic and appropriate results can be obtained. For example, if there is a lot of negative emotional feedback, the demand forecast volume can be reduced. The inputs are the prediction results and emotional data, and the output is the adjusted prediction results.

[0514] Step 6:

[0515] The mobile terminal or wearable device provides the adjusted prediction results to the user in real time. The user can check the information through the device and make corrections as needed. Corrections are made via voice or touch input, and the data is sent back to the server. The input is the adjusted prediction results, and the output is the corrected prediction results or user feedback.

[0516] Step 7:

[0517] The server automatically executes the final action based on the user-modified information, such as sending the final inventory order list to suppliers or notifying employees of shift schedules. The input is the modified prediction results and feedback, and the output is the specific action to be taken.

[0518] Through the above steps, the server, mobile terminal, wearable device, and user can work together to optimize the business operations of a physical store and improve user satisfaction.

[0519] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0520] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0521] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0522] [Second embodiment]

[0523] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0524] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0525] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0526] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0527] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0528] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0529] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0530] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0531] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0532] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0533] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0534] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0535] This invention relates to a system for optimizing specific business operations, and aims to improve business efficiency by applying AI models to mainly utilize customer purchasing information, inventory data, waste data, and employee shift information. The system consists of three main components: a server, a terminal, and a user.

[0536] Automated ordering system

[0537] Server: In the ordering system, the server first collects inventory status, past order data, and waste data. Next, the server preprocesses this data and fills in missing and outlier values. Based on the preprocessed data, the server uses an AI model to forecast demand and calculate the next order quantity.

[0538] For example, if an analysis of data from the past year predicts that demand for a particular product will increase next week, the server automatically generates an appropriate order quantity based on that prediction. After the user checks this order list and makes any necessary corrections, the server automatically sends the final list to the supplier.

[0539] Dynamic Pricing System

[0540] Server: In this system, the server collects customer purchase history and store visits by time of day, and pre-processes this data. Based on the pre-processed data, the server uses an AI model to analyze the relationship between demand and price. Next, it sets the optimal price based on demand and automatically distributes it to the POS system.

[0541] For example, if past purchase history shows that demand is highest between 3:00 PM and 6:00 PM on weekdays, you can set the price to be 10% higher during that time period, which will reduce unsold items and maximize profits.

[0542] Automatic shift creation

[0543] Server: In the shift creation system, the server collects employee shift history, desired vacation time, and store visit predictions. It then preprocesses this data and uses an AI model to predict optimal shifts based on the number of employees and peak hours. The predicted shift schedule is automatically generated and can be reviewed and modified by the user.

[0544] For example, if a high number of customers is predicted on a particular weekend, the system generates a shift schedule that automatically allocates the necessary number of employees during that time period. The user can check the generated schedule, correct any problems, and automatically distribute it to employees.

[0545] Service and product demand forecasting

[0546] Server: In a demand forecasting system, the server collects and preprocesses customer purchase history and sales data. Based on the preprocessed data, the server uses an AI model to predict future demand. Based on the prediction results, it generates a list of recommended promotional items and inventory replenishment and presents it to the user.

[0547] For example, if past data predicts that a particular product will be popular in the next month, the product will be listed as a promotion target. Users can check the list and make appropriate promotion plans and inventory replenishment.

[0548] In this way, the invention will reduce food waste, alleviate labor shortages, and achieve sustainable service provision through appropriate pricing.

[0549] The processing flow will be explained below.

[0550] Automated ordering system processing

[0551] Step 1:

[0552] Server: Collects inventory status, past order data, and disposal data from the database.

[0553] Every day at 2:00 AM, the server collects inventory status, order data, and disposal data from the database for the past year.

[0554] Step 2:

[0555] Server: Preprocesses the collected data and imputes missing or outlier values.

[0556] After collecting the data, the server performs preprocessing, imputing missing values ​​with the average value and removing outliers.

[0557] Step 3:

[0558] Server: Preprocessed data is input into the AI ​​model to generate demand forecasts.

[0559] The server passes the pre-processed data to an AI model to generate demand forecasts for the next week.

[0560] Step 4:

[0561] Server: Calculates the optimal order quantity based on the demand forecast results.

[0562] The server calculates the optimal order quantity for each product based on the demand forecast results output by the AI ​​model.

[0563] Step 5:

[0564] Server: Automatically generate an order list based on the optimal order quantity.

[0565] The server creates an order list based on the calculated order quantity.

[0566] Step 6:

[0567] User: Check the generated order list and modify it as necessary.

[0568] The user checks the order list on the system and corrects the quantity if necessary.

[0569] Step 7:

[0570] Server: Automatically sends the confirmed and corrected order list to the supplier.

[0571] The server receives the order list that the user has confirmed and modified, and automatically sends it to the specified supplier.

[0572] Dynamic pricing system processing

[0573] Step 1:

[0574] Server: Collects customer purchase history and store visit data by time of day.

[0575] Every night, the server collects customer purchase history and time-of-day visit data from the database.

[0576] Step 2:

[0577] Server: Preprocesses the collected data and removes outliers.

[0578] The server performs preprocessing after data collection to normalize or remove outliers.

[0579] Step 3:

[0580] Server: Inputs the preprocessed data into the AI ​​model and analyzes the relationship between demand and price.

[0581] The server passes the pre-processed data to an AI model that analyzes the relationship between demand and price.

[0582] Step 4:

[0583] Server: Sets the price for the next day based on the analysis results.

[0584] The server uses the analysis results output by the AI ​​model to set the optimal price for the next day.

[0585] Step 5:

[0586] Server: Automatically distributes the set price information to the POS system.

[0587] The server sends the set price information to the POS system, which updates it automatically.

[0588] Process automatic shift creation

[0589] Step 1:

[0590] Server: Collects employee shift history, desired vacation time, and store visit forecast data.

[0591] At the end of each month, the server collects employee shift history, desired vacation data, and store visit forecast data from the database.

[0592] Step 2:

[0593] Server: Preprocesses the collected data and imputes missing or outlier values.

[0594] After data collection, the server performs preprocessing, imputing missing values ​​and removing outliers.

[0595] Step 3:

[0596] Server: Based on the pre-processed data, the AI ​​model predicts the optimal shift.

[0597] The server inputs the pre-processed data into an AI model to predict the optimal shift schedule.

[0598] Step 4:

[0599] Server: Automatically generates predicted shift schedules.

[0600] The server automatically generates shift schedules based on the prediction results output by the AI ​​model.

[0601] Step 5:

[0602] User: Review the generated shift schedule and modify it if necessary.

[0603] Users can check the shift schedule on the system and modify it if necessary.

[0604] Step 6:

[0605] Server: Automatically distributes confirmed shift schedules to employees.

[0606] The server receives the shift schedule that the user has confirmed and modified and automatically distributes it to all employees.

[0607] Processing demand forecasts for services and products

[0608] Step 1:

[0609] Server: Collects customer purchase history and sales data.

[0610] The server collects customer purchase history and sales data from the database every day at 11:00 p.m.

[0611] Step 2:

[0612] Server: Preprocesses the collected data and imputes missing or outlier values.

[0613] After data collection, the server performs preprocessing, imputing missing values ​​and removing outliers.

[0614] Step 3:

[0615] Server: Inputs the pre-processed data into the AI ​​model to predict future demand.

[0616] The server passes the pre-processed data to an AI model to predict future demand.

[0617] Step 4:

[0618] Server: Based on the prediction results, a list of recommended promotion targets and inventory replenishment is generated.

[0619] The server automatically generates a list of recommended promotional targets and inventory replenishment based on the prediction results output by the AI ​​model.

[0620] Step 5:

[0621] User: Review the generated recommendation list and incorporate it into promotion planning and inventory replenishment.

[0622] The user checks the recommended list and implements promotion plans and inventory replenishment as necessary.

[0623] Example 1

[0624] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0625] In conventional business systems, customer purchasing information, inventory data, disposal data, and employee shift information are often managed separately, making it difficult to integrate and efficiently utilize this data. Furthermore, business processes such as demand forecasting, pricing, and automatic shift generation are performed manually, requiring a great deal of time and effort, making it difficult to improve the efficiency and accuracy of business operations. Aiming to solve these problems, the present invention provides a system for optimizing specific business operations.

[0626] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0627] In this invention, the server includes means for collecting customer purchasing information, inventory data, disposal data, and employee shift information, means for preprocessing the collected data and complementing missing values ​​and outliers, means for using a generative AI model based on the preprocessed data to perform demand forecasting, optimizing inventory order quantities, optimizing pricing, and automatically generating employee shifts, means for presenting necessary information to a user based on the prediction results and generated data and for confirmation and correction, and means for automatically executing necessary actions based on the information confirmed and corrected by the user. This enables efficient management of customer purchasing information, inventory data, disposal data, and employee shift information, automating business processes such as demand forecasting, inventory ordering, pricing, and shift creation, and improving business efficiency and accuracy.

[0628] Key Word Definitions

[0629] "Customer purchasing information" refers to data that records the type, quantity, purchase date, etc. of products that a customer has purchased in the past.

[0630] "Inventory data" refers to data for managing the quantity of products currently in stock, storage locations, and inventory history.

[0631] "Disposal data" refers to data that records the quantity of products that have been discarded due to reasons such as expiration or poor quality, the reason for disposal, and the date and time of disposal.

[0632] "Employee shift information" refers to data necessary for labor management, such as employees' working hours, desired holidays, and shift history.

[0633] "Missing values" refers to a state in which data that should be present in a dataset is missing.

[0634] An "outlier" is a value that is significantly different from other data in a data set and is typically treated as an error or outlier.

[0635] A "generative AI model" is a model that uses artificial intelligence technology to learn specific patterns from data and automatically perform tasks such as prediction and classification.

[0636] "Demand forecasting" is the process of predicting future demand based on past data, and refers to predicting sales quantities, usage amounts, etc.

[0637] "Inventory order optimization" is the process of determining optimal inventory levels based on demand forecasts to avoid shortages and excess inventory.

[0638] "Pricing optimization" is the process of setting optimal selling prices taking into account demand and market conditions, with the goal of maximizing profits.

[0639] "Automatic employee shift generation" is the process of automatically creating employee work schedules, taking into account business demands and employee preferences.

[0640] "Means for automatically taking necessary actions based on information confirmed and corrected by the user" refers to a process for automatically taking business actions such as placing orders, changing prices, and notifying shifts based on data confirmed and corrected by the user.

[0641] MODE FOR CARRYING OUT THE INVENTION

[0642] This invention relates to a system for optimizing specific business operations, and aims to improve business efficiency by applying generative AI models primarily using customer purchasing information, inventory data, waste data, and employee shift information. This system consists of three main components: a server, a terminal, and a user.

[0643] Hardware and software used

[0644] The server is built on AWS (a common cloud computing service) and uses MySQL as the database. Python and libraries such as Pandas and Scikit-learn are used for data preprocessing and analysis. AI frameworks such as TensorFlow, Keras, XGBoost, and Prophet are mainly used to implement the generative AI model. A general sales management system is assumed as the POS system.

[0645] System example and operation

[0646] Automated ordering system

[0647] Server: The server periodically collects inventory status, past order data, and waste data from terminals and point-of-sale (POS) systems, and preprocesses this data using Python and Pandas. It imputes missing and outlier values ​​and inputs the data into a generative AI model (e.g., TensorFlow) to perform demand forecasting. It automatically generates an order list based on the forecast results and presents it to the user. The user can review and modify the order list, and the final list is automatically sent from the server to the supplier.

[0648] Example: Based on data from the past year, the order quantity for the next week, when demand is predicted to increase, is automatically calculated and an order list is created.

[0649] Dynamic Pricing System

[0650] Server: The server collects and preprocesses customer purchase history and store visit status by time of day in real time. This data is input into a generative AI model (e.g., XGBoost) to analyze the relationship between demand and price. The optimal price is calculated from the analysis results and automatically distributed to the sales management system.

[0651] Example: If analysis shows that demand is high between 3pm and 6pm on weekdays, the price during that time period will be increased by 10%.

[0652] Automatic shift creation

[0653] Server: The server collects and preprocesses employee shift history, desired vacation time, and store visit prediction information. This data is input into a generative AI model (e.g., Keras) to predict and automatically generate employee shift schedules. Users can check and modify the generated shift schedules. The final shift schedules are automatically distributed from the server to employees.

[0654] Example: If store visits are predicted to be high on a particular weekend, create a shift that places more employees at that time.

[0655] Service and product demand forecasting

[0656] Server: The server periodically collects customer purchase history and sales data, and after preprocessing, inputs it into a generative AI model (e.g., Prophet) to predict future demand. Based on the prediction results, it generates a list of recommended promotional items and inventory replenishment and presents it to the user. The user checks this list and takes the necessary actions.

[0657] Example: If historical data predicts that a particular product will be popular next month, target that product for promotion.

[0658] Prompt Sentence Examples

[0659] "Please forecast demand for the next month based on data from the past year, and propose periods when demand will be high and product promotions. Please also let us know specific promotion ideas."

[0660] In this way, the present invention is a system that uses a generative AI model to integrate and analyze a variety of business data and automatically implement optimized business processes, thereby improving business efficiency and supporting highly accurate decision-making.

[0661] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0662] Processing steps of specific business operation optimization system

[0663] Automated ordering system processing flow

[0664] Step 1: Data collection

[0665] Server: The server periodically collects inventory status, past order data, and waste data from terminals and POS systems. Raw data obtained from each terminal is sent to the server as input. The output is consolidated data stored in a database within the server. As a concrete example, an automatic data collection script is executed every night.

[0666] Step 2: Data Preprocessing

[0667] Server: The collected data is preprocessed using Python and Pandas. The collected raw data is used as input. Missing values ​​and outliers are imputed, and the output is formatted as time series data. For example, missing values ​​are imputed with the past average value, and outliers with the median value.

[0668] Step 3: Demand forecasting using AI models

[0669] Server: The preprocessed data is input into a generative AI model (e.g., TensorFlow) to generate a demand forecast. The forecast model takes into account trends and seasonal fluctuations based on past data. The preprocessed data is used as input, and a forecast of future demand is output. For example, if demand for next week is predicted to be 200 units, the server outputs that number.

[0670] Step 4: Generate an order list

[0671] Server: Automatically generates an order list based on the forecast results. Based on the forecasted demand, the optimal order quantity is calculated taking into account the safety stock amount, and compiled in list format. The demand forecast results are used as input, and the output is an order list. As a specific example, if the forecasted demand is 200 units, an order list for 220 units is generated taking into account the safety stock amount.

[0672] Step 5: Check and modify the order list

[0673] User: The user checks the generated order list in a dedicated interface and modifies it if necessary. The automatically generated order list is used as input, and a modified final list is generated as output. For example, if the user feels that there are too many of a particular product in this list, the order quantity can be adjusted by the user.

[0674] Step 6: Send the order list to your suppliers

[0675] Server: Automatically sends the final order list, which has been reviewed and modified by the user, to the supplier. It uses the modified order list as input and generates the order instructions that are sent to the supplier as output, for example, via email or electronic data interchange (EDI).

[0676] Dynamic pricing system processing flow

[0677] Step 1: Data collection

[0678] Server: The server collects customer purchase history and store visit status by time of day in real time. The input is raw data obtained from the POS system, and the output is purchase data stored in a database within the server. For example, the latest data is obtained from the POS system every hour.

[0679] Step 2: Data Preprocessing

[0680] Server: Cleanses the collected data and removes outliers using Python and Scikit-learn. The input is raw data collected from the POS system, and the output is a clean dataset. For example, the format of dates and times is standardized.

[0681] Step 3: Analyze the relationship between demand and price

[0682] Server: The preprocessed data is input into an AI model (e.g., XGBoost) to analyze the relationship between demand and price. The input is the preprocessed data, and the output is the analysis result. For example, demand forecasting is performed using the relationship between past price fluctuations and demand.

[0683] Step 4: Set the optimal price

[0684] Server: Sets the optimal price based on the analysis results. The analysis results are used as input and new pricing is generated as output. For example, the price is set to increase by 10% during times of high demand.

[0685] Step 5: Distributing price information to POS systems

[0686] Server: Calculated price information is sent to the POS system in real time, automatically updating the store's prices. The input is the new price setting, and the output is the price information reflected in the POS system. For example, the new price information is immediately applied to the POS system and displayed to the customer.

[0687] Process flow for automatic shift creation

[0688] Step 1: Data collection

[0689] Server: The server collects employee shift history, desired vacation time, and store visit prediction information. The employee database and store visit prediction data are used as input, and the output is the integrated data stored in the server. For example, the latest data is obtained every Sunday.

[0690] Step 2: Data Preprocessing

[0691] Server: Cleanses the collected data and processes missing and outliers using Python and Pandas. The input is raw data, and the output is a clean dataset. For example, missing values ​​are imputed with the mean and outliers are removed.

[0692] Step 3: Shift prediction using AI models

[0693] Server: Input the preprocessed data into an AI model (e.g., Keras) to predict employee shifts. The input is the preprocessed data, and the output is the predicted shift schedule. For example, it allocates the required number of employees according to peak hours.

[0694] Step 4: Generate a shift schedule

[0695] Server: Automatically generates shift schedules based on the prediction results. The input is the prediction results, and the output is the generated shift schedule. As a specific example, shifts are adjusted according to the predicted number of customers.

[0696] Step 5: Check and modify the shift schedule

[0697] User: The user checks the generated shift schedule and modifies it if necessary. The input is the automatically generated shift schedule, and the output is the final modified schedule. For example, if a user requests sudden leave, the shift can be covered by another employee.

[0698] Step 6: Distribute shift schedules to employees

[0699] Server: Automatically distributes the final shift schedule to employees. The input is the revised shift schedule, and the output is the shift information sent to employees. For example, it can be sent via email or a dedicated app.

[0700] Service / product demand forecasting process flow

[0701] Step 1: Data collection

[0702] Server: The server periodically collects customer purchase history and sales data. The input is raw data obtained from the POS system, and the output is sales data stored in a database on the server. For example, data is collected at the end of each month.

[0703] Step 2: Data Preprocessing

[0704] Server: Cleanses the collected data and processes missing values ​​and outliers using Python and Pandas. The input is raw data, and the output is a clean dataset. For example, missing values ​​are imputed with the average value of the previous month, and outliers are removed.

[0705] Step 3: Demand forecasting using AI models

[0706] Server: Inputs the preprocessed data into a generative AI model (e.g., Prophet) to predict future demand. The input is the preprocessed data, and the output is the prediction result. For example, a specific product is expected to be in high demand next month.

[0707] Step 4: Generate a recommendation list

[0708] Server: Based on the prediction results, a list of recommendations for promotions and inventory replenishment is generated. The input is the prediction results, and the output is the generated recommendation list. As a concrete example, it lists products with increasing demand.

[0709] Step 5: Review and modify the recommendation list

[0710] User: The user reviews the generated recommendation list and modifies it if necessary. The input is the automatically generated recommendation list, and the output is the modified list. For example, the user determines that a critical item needs to be restocked.

[0711] Step 6: Planning promotions and replenishing inventory

[0712] User: After reviewing the recommendation list, create promotion plans or replenish inventory based on the list. The input is the revised recommendation list, and the output is the implemented promotions or replenished inventory. For example, target specific products for a special campaign.

[0713] (Application example 1)

[0714] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0715] In traditional brick-and-mortar stores, inventory management, pricing, demand forecasting, and employee shift planning are often performed manually, resulting in problems such as inefficiency and human error in each process. Furthermore, there was a lack of means to grasp trends in customer purchasing behavior in real time and respond quickly, making appropriate inventory management and pricing difficult. Furthermore, it was difficult to implement automatic ordering based on forecasts or dynamic pricing to prevent unsold items, making it difficult to develop effective sales strategies. It is essential to solve these problems and improve the overall efficiency of store operations.

[0716] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0717] In this invention, the server includes: means for collecting customer purchasing information, inventory data, disposal data, and employee shift information; means for preprocessing the collected data and complementing missing and outlier values; means for using an AI model based on the preprocessed data to forecast demand, optimize inventory order quantities, optimize pricing, and automatically generate employee shifts; means for presenting necessary information to a user based on the forecast results and generated data and allowing the user to confirm and correct it; means for automatically executing necessary actions based on the information confirmed and corrected by the user; means for providing inventory management functions in real time and issuing alerts to prevent stockouts and excess inventory; means for dynamically setting product prices based on purchase history and the number of customers by time period to prevent unsold items and maximize profits; means for automatically creating optimal shift schedules based on employee shift history and desired vacation days; and means for predicting future demand based on customer purchase history and sales data and generating recommended lists of promoted products and inventory replenishment. This allows for centralized management of various processes related to store operations, enabling efficient and accurate inventory management, pricing, demand forecasting, and shift creation.

[0718] "Customer purchasing information" is data including the type and quantity of products purchased by customers who visit a store, the date and time of purchase, and the attributes of the purchaser.

[0719] "Inventory data" refers to information relating to the current quantity of products held in the store, storage locations, replenishment status, and the like.

[0720] "Disposal data" is data indicating the type, quantity, and date and time of disposal of products that have been discarded due to the expiration of their sales period or deterioration of quality.

[0721] "Employee shift information" is data including an employee's working date and time, working hours, vacation days, desired work shifts, etc.

[0722] "Preprocessing" is the process of converting collected data into a format suitable for analysis and model training, and filling in missing or outlier values.

[0723] An "AI model" is a mathematical model that uses artificial intelligence technologies such as machine learning and deep learning to make useful predictions and classifications from data.

[0724] "Demand forecasting" is the process of predicting future demand for a product based on past data.

[0725] "Inventory order quantity optimization" is the process of preventing excess inventory and out-of-stock by predicting product demand and calculating appropriate order quantities based on that.

[0726] "Pricing optimization" is the process of adjusting the price of a product to an optimal value based on demand data, etc.

[0727] "Automatic generation of employee shifts" is the process of automatically creating optimal shift schedules based on employees' shift history, desired working hours, and store business demands.

[0728] "Prediction results and generated data" refers to information such as demand forecast results, optimal order quantities, appropriate prices, and shift schedules generated by AI models.

[0729] The "inventory management function" monitors inventory status in real time and reduces the risk of stockouts and excess inventory.

[0730] "Dynamic setting" means that values ​​and parameters can be flexibly changed according to real-time data and the current situation.

[0731] "Promotional Products" means selected products that are specially promoted or discounted for promotional purposes.

[0732] A "recommended list" is a list generated by the system to encourage specific actions or responses.

[0733] This invention is built as a system for optimizing specific business operations. The system consists of three main components: a server, a terminal, and a user.

[0734] Inventory management function

[0735] The server first collects customer purchasing information, inventory data, waste data, and employee shift information. This data is obtained from databases and POS systems. The server then preprocesses the data and fills in missing and outlier values. Specifically, it cleanses the data using data analysis libraries such as Pandas. Using the preprocessed data, it trains a demand forecasting model using AI libraries such as TensorFlow and Keras. This model predicts future inventory demand and calculates appropriate order quantities.

[0736] Dynamic Pricing Feature

[0737] The server also analyzes purchase history and store visits by time of day to determine optimal pricing based on demand. This data is also preprocessed and fed into the AI ​​model, which analyzes the relationship between demand and price and calculates the most profitable price. This price information is automatically sent to the POS system, where prices are adjusted in real time.

[0738] Automatic shift creation

[0739] The server automatically generates optimal shift schedules based on employees' shift history, desired holidays, and predicted store visits. Employee information and store visit prediction data are fed into an AI model to calculate the optimal shift schedule. Users can check and modify the generated schedule, and the final shift schedule is distributed to employees.

[0740] Demand forecasting and promotions

[0741] In addition, the server collects customer purchase history and sales data to predict future demand. Based on this prediction, it targets specific products for promotion and generates a list of recommended inventory replenishment. Specifically, if the AI ​​model predicts that a specific product will be popular in the next month, it will list that product as a promotion target.

[0742] Hardware and software used

[0743] The hardware used is a server for database hosting and AI model training, and a smartphone as a client terminal. The software used is Pandas (data collection and preprocessing) and TensorFlow / Keras (demand forecasting AI model construction and training).

[0744] Examples of specific examples and prompts

[0745] For example, if a particular product is predicted to be popular in the coming month, the product may be listed for promotion and the store may be notified.

[0746] Example prompt sentence:

[0747] "Using past sales data, the AI ​​model predicts which products will be in high demand for the next month, and creates a promotion plan for those products."

[0748] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0749] Step 1:

[0750] The server collects customer purchasing information, inventory data, disposal data, and employee shift information.

[0751] These data are obtained from databases and POS systems. The collected data is first converted into a data frame using the Pandas library, and missing and outlier values ​​are imputed (data cleansing).

[0752] Input: Customer purchasing information, inventory data, disposal data, employee shift information

[0753] Output: Preprocessed data frame

[0754] Step 2:

[0755] The server trains the AI ​​model based on the preprocessed data.

[0756] Build and train a demand forecasting model using TensorFlow and Keras libraries. Historical purchase and inventory data are used as input data for the model to generate demand forecasts.

[0757] Input: Preprocessed data frame

[0758] Output: A trained demand forecasting model

[0759] Step 3:

[0760] The server uses the trained AI model to predict future inventory demand and calculate appropriate order quantities.

[0761] Past purchasing data is input into the forecasting model to predict demand for the next week or month. Based on the forecast results, the optimal inventory order quantity is calculated.

[0762] Inputs: Trained demand forecasting model, historical purchase data

[0763] Output: Future inventory demand forecast, order quantity

[0764] Step 4:

[0765] The server automatically generates an order list based on the demand forecast results and the order quantity, and presents it to the user.

[0766] The generated order list is displayed on the terminal for the user to review and modify, and after the user makes the appropriate modifications, the information is automatically sent to the supplier.

[0767] Input: Future inventory demand forecast, order quantity

[0768] Output: Order list, order data to suppliers

[0769] Step 5:

[0770] The server analyzes purchase history and store visits by time of day, and sets optimal prices based on demand.

[0771] The AI ​​model analyzes the relationship between demand and price to calculate the optimal price, which is then automatically sent to the POS system.

[0772] Input: Purchase history, store visits by time period

[0773] Output: Best price, price information to POS system

[0774] Step 6:

[0775] The server automatically generates an optimal shift schedule based on employee shift history, desired holidays, and predicted store visits, and presents it to the user.

[0776] The AI ​​model calculates the optimal shift schedule, which the user can then review and modify. The modified schedule is then distributed to employees.

[0777] Input: Shift history, desired leave, store visit forecast

[0778] Output: Shift schedule, information distributed to employees

[0779] Step 7:

[0780] The server predicts future demand based on customer purchase history and sales data, generates a recommended list of promotional items and inventory replenishment, and presents it to the user.

[0781] Based on the generated recommendation list, the user plans promotions and stock replenishment.

[0782] Input: Purchase history, sales data

[0783] Output: Demand forecast, recommendation list, promotional products

[0784] The above are the specific processing steps for carrying out the present invention.

[0785] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0786] This invention relates to a system for optimizing specific business operations, primarily using customer purchasing information, inventory data, waste data, and employee shift information to apply AI models to improve business efficiency. Furthermore, by combining it with an emotion engine that recognizes user emotions, even more accurate predictions and optimization can be achieved. This system consists of three main components: a server, a terminal, and a user.

[0787] Automated ordering system

[0788] Server: In the ordering system, the server first collects inventory status, past order data, and waste data. Next, the server preprocesses this data and fills in missing and outlier values. Based on the preprocessed data, the server uses an AI model to forecast demand and calculate the next order quantity. It also uses an emotion engine to incorporate user feedback and adjust the optimal order quantity.

[0789] For example, if an analysis of data from the past year predicts that demand for a particular product will increase next week, the system automatically generates an appropriate order quantity based on that prediction. After the user reviews this order list and the emotion engine analyzes the user's reactions and makes any necessary adjustments, the server automatically sends the final list to the supplier.

[0790] Dynamic Pricing System

[0791] Server: In this system, the server collects customer purchase history and store visits by time of day, and pre-processes this data. Based on the pre-processed data, the server uses an AI model to analyze the relationship between demand and price. Next, the server sets the optimal price based on demand, and an emotion engine incorporates user feedback to adjust the price setting, which is then automatically distributed to the POS system.

[0792] For example, if past purchasing history shows that demand is high between 3:00 PM and 6:00 PM on weekdays, the price can be increased by 10% during that time period, reducing unsold items and maximizing profits. The emotion engine analyzes user reactions and readjusts prices based on feedback.

[0793] Automatic shift creation

[0794] Server: In the shift creation system, the server collects employee shift history, desired vacation time, and store visit predictions. It then preprocesses this data and uses an AI model to predict optimal shifts based on the number of employees and peak hours. The predicted shift schedule is automatically generated and confirmed by the user. An emotion engine also analyzes the user's reactions and adjusts the shift schedule as necessary.

[0795] As a specific example, if a high number of customers is predicted on a particular weekend, the system will automatically generate a shift schedule that allocates the necessary number of employees during that time period. The user can check the generated schedule, and the emotion engine will analyze the user's reactions, correct any problems, and automatically distribute the schedule to employees.

[0796] Service and product demand forecasting

[0797] Server: In a demand forecasting system, the server collects and preprocesses customer purchase history and sales data. Based on the preprocessed data, the server uses an AI model to predict future demand. Based on the prediction results, it generates a list of recommended promotional items and inventory replenishment and presents it to the user. An emotion engine analyzes user reactions and adjusts the recommended list.

[0798] For example, if past data predicts that a particular product will be popular in the next month, the product will be listed as a promotion target. Users can check the list, and the emotion engine will analyze their reactions and make appropriate adjustments, such as planning promotions and replenishing inventory.

[0799] In this way, the invention will realize sustainable service provision through food waste reduction, labor shortage mitigation, and appropriate pricing. In addition, by integrating an emotion engine, more accurate prediction and optimization will be possible, leading to improved user satisfaction.

[0800] The processing flow will be explained below.

[0801] Automated ordering system processing

[0802] Step 1:

[0803] Server: Collects inventory status, past order data, and disposal data from the database.

[0804] Every day at 2:00 AM, the server collects inventory status, order data, and disposal data from the database for the past year.

[0805] Step 2:

[0806] Server: Preprocesses the collected data and imputes missing or outlier values.

[0807] After collecting the data, the server performs preprocessing, imputing missing values ​​with the average value and removing outliers.

[0808] Step 3:

[0809] Server: Preprocessed data is input into the AI ​​model to generate demand forecasts.

[0810] The server passes the pre-processed data to an AI model to generate demand forecasts for the next week.

[0811] Step 4:

[0812] Server: Calculates the optimal order quantity based on the demand forecast results.

[0813] The server calculates the optimal order quantity for each product based on the demand forecast results output by the AI ​​model.

[0814] Step 5:

[0815] Server: Automatically generate an order list based on the optimal order quantity.

[0816] The server creates an order list based on the calculated order quantity.

[0817] Step 6:

[0818] User: Check the generated order list and modify it as necessary.

[0819] The user checks the order list on the system and corrects the quantity if necessary.

[0820] Step 7:

[0821] Emotion Engine: Analyzes user reactions and suggests adjustments to the order list.

[0822] The emotion engine analyzes user reactions in real time and suggests adjustments to order quantities.

[0823] Step 8:

[0824] Server: Automatically sends the confirmed and corrected order list to the supplier.

[0825] The server receives the order list that the user has confirmed and modified, and automatically sends it to the specified supplier.

[0826] Dynamic pricing system processing

[0827] Step 1:

[0828] Server: Collects customer purchase history and store visit data by time of day.

[0829] Every night, the server collects customer purchase history and time-of-day visit data from the database.

[0830] Step 2:

[0831] Server: Preprocesses the collected data and removes outliers.

[0832] The server performs preprocessing after data collection to normalize or remove outliers.

[0833] Step 3:

[0834] Server: Inputs the preprocessed data into the AI ​​model and analyzes the relationship between demand and price.

[0835] The server passes the pre-processed data to an AI model that analyzes the relationship between demand and price.

[0836] Step 4:

[0837] Server: Sets the price for the next day based on the analysis results.

[0838] The server uses the analysis results output by the AI ​​model to set the optimal price for the next day.

[0839] Step 5:

[0840] Emotion Engine: Analyzes user reactions and suggests pricing adjustments.

[0841] The sentiment engine analyzes user reactions in real time and suggests pricing adjustments.

[0842] Step 6:

[0843] Server: Automatically distributes the set price information to the POS system.

[0844] The server sends the set price information to the POS system, which updates it automatically.

[0845] Process automatic shift creation

[0846] Step 1:

[0847] Server: Collects employee shift history, desired vacation time, and store visit forecast data.

[0848] At the end of each month, the server collects employee shift history, desired vacation data, and store visit forecast data from the database.

[0849] Step 2:

[0850] Server: Preprocesses the collected data and imputes missing or outlier values.

[0851] After data collection, the server performs preprocessing, imputing missing values ​​and removing outliers.

[0852] Step 3:

[0853] Server: Based on the pre-processed data, the AI ​​model predicts the optimal shift.

[0854] The server inputs the pre-processed data into an AI model to predict the optimal shift schedule.

[0855] Step 4:

[0856] Server: Automatically generates predicted shift schedules.

[0857] The server automatically generates shift schedules based on the prediction results output by the AI ​​model.

[0858] Step 5:

[0859] User: Review the generated shift schedule and modify it if necessary.

[0860] Users can check the shift schedule on the system and modify it if necessary.

[0861] Step 6:

[0862] Emotion engine: Analyzes user reactions and suggests shift schedule adjustments.

[0863] The emotion engine analyzes user reactions in real time and suggests adjustments to shift schedules.

[0864] Step 7:

[0865] Server: Automatically distributes confirmed shift schedules to employees.

[0866] The server receives the shift schedule that the user has confirmed and modified and automatically distributes it to all employees.

[0867] Processing demand forecasts for services and products

[0868] Step 1:

[0869] Server: Collects customer purchase history and sales data.

[0870] The server collects customer purchase history and sales data from the database every day at 11:00 p.m.

[0871] Step 2:

[0872] Server: Preprocesses the collected data and imputes missing or outlier values.

[0873] After data collection, the server performs preprocessing, imputing missing values ​​and removing outliers.

[0874] Step 3:

[0875] Server: Inputs the pre-processed data into the AI ​​model to predict future demand.

[0876] The server passes the pre-processed data to an AI model to predict future demand.

[0877] Step 4:

[0878] Server: Based on the prediction results, a list of recommended promotion targets and inventory replenishment is generated.

[0879] The server automatically generates a list of recommended promotional targets and inventory replenishment based on the prediction results output by the AI ​​model.

[0880] Step 5:

[0881] User: Review the generated recommendation list and incorporate it into promotion planning and inventory replenishment.

[0882] The user checks the recommended list and implements promotion plans and inventory replenishment as necessary.

[0883] Step 6:

[0884] Emotion engine: Analyzes user reactions and suggests adjustments to the recommendation list.

[0885] The emotion engine analyzes user reactions in real time and makes suggestions to adjust the recommendation list.

[0886] In this way, the invention will realize sustainable service provision through food waste reduction, labor shortage mitigation, and appropriate pricing. In addition, by integrating an emotion engine, more accurate prediction and optimization will be possible, leading to improved user satisfaction.

[0887] Example 2

[0888] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0889] In conventional business operation systems, customer purchasing information, inventory data, disposal data, and employee shift information are managed separately, making data integration and management cumbersome and making efficient operation difficult. Furthermore, predictions and optimization based on this data are often performed manually, limiting the efficiency of operations. Furthermore, because feedback that takes user emotions into account is not reflected, optimization results may not meet on-site needs. Therefore, a system that can solve these issues all at once is needed.

[0890] The identification processing by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting customer purchasing information, inventory data, disposal data, and employee shift information, means for preprocessing the collected data and complementing missing values ​​and outliers, means for performing demand forecasting using a generative AI model based on the preprocessed data, means for optimizing order quantities, optimizing pricing, and automatically generating employee shifts based on the generated demand forecast, means for obtaining user feedback using a sentiment analysis engine based on the prediction results and the generated data and adjusting necessary information, and means for automatically executing necessary actions based on information confirmed and corrected by the user. This enables integrated data management and efficient business operations, and realizes highly accurate predictions and optimization that reflect user feedback.

[0891] "Customer purchasing information" refers to data such as the products purchased by the customer, the purchase date and time, and the quantity purchased.

[0892] "Inventory data" refers to data including information such as product inventory quantity, date of receipt, and storage location.

[0893] "Disposal data" refers to data that includes information such as the quantity of inventory items to be discarded, the reason for disposal, and the date and time of disposal.

[0894] "Employee shift information" refers to data such as employees' working days, working hours, and desired holidays.

[0895] "Means of collection" refers to the methods and equipment used to obtain the necessary information from databases and sensors.

[0896] "Preprocessing means" refers to a method for completing missing data values, eliminating outliers, and converting the data into a format suitable for analysis.

[0897] A "generative AI model" refers to a model that includes algorithms for demand forecasting and optimization from data using techniques such as machine learning and deep learning.

[0898] "Means for demand forecasting" refers to methods and systems that use AI models to estimate future demand.

[0899] "Order quantity optimization" refers to the process of determining the optimal product order quantity based on demand forecasts.

[0900] "Pricing optimization" refers to the process of determining the most effective price by taking into account the balance between supply and demand.

[0901] "Automatic employee shift generation" refers to a system that automatically creates work schedules based on employee preferences and business needs.

[0902] An "emotion analysis engine" refers to a technology that analyzes emotions from user input, facial expressions, and voice, and reflects that feedback in the system.

[0903] "Means for adjusting required information" refers to a method or system that adjusts predictions or optimization results based on user feedback obtained using a sentiment analysis engine.

[0904] "Means for automatically executing necessary actions" refers to a method or system whereby the system automatically executes the relevant business process based on the information confirmed or corrected by the user.

[0905] This invention relates to a system for optimizing specific business operations, which utilizes customer purchasing information, inventory data, waste data, and employee shift information to apply a generative AI model to improve business efficiency. Furthermore, by combining it with a sentiment analysis engine, it achieves even more accurate prediction and optimization.

[0906] System configuration

[0907] This system consists of three main components: the server, the terminal, and the user. Each component will be explained in detail below.

[0908] Data collection and preprocessing

[0909] The server first collects customer purchasing information, inventory data, disposal data, and employee shift information. A relational database (e.g., MySQL or PostgreSQL) is used. After collecting the data, the server preprocesses the data and fills in missing and outlier values. Specifically, data cleaning is performed using the Python pandas library.

[0910] Demand forecasting and optimization

[0911] Based on the preprocessed data, the server uses a generative AI model (e.g., TensorFlow, PyTorch) to generate demand forecasts. Based on the generated demand forecasts, the server optimizes order quantities, optimizes pricing, and automatically generates employee shifts. For example, it uses a time series analysis model to forecast demand.

[0912] Sentiment analysis and feedback adjustment

[0913] The user checks the generated prediction and optimization results. Here, a sentiment analysis engine (e.g., OpenAI's GPT-4 model) analyzes the user's feedback and makes adjustments as necessary. Specifically, the user can check and correct the results using a dashboard provided in the web application.

[0914] Autorun and Notifications

[0915] The server automatically executes the necessary actions based on the information confirmed and corrected by the user. For example, if an order list is confirmed, the server automatically sends the order to the supplier using an API. It also distributes price information to the POS system.

[0916] Specific examples

[0917] A concrete example of how this works is shown below:

[0918] 1. Automated ordering system:

[0919] After analyzing data from the past year, it predicts that demand for a particular product will increase next week. Based on this prediction, an order list is generated and confirmed by the user. A sentiment analysis engine analyzes the user's reaction and adjusts the order quantity to the optimum. Finally, the server automatically sends the order list to the supplier.

[0920] 2. Dynamic Pricing System:

[0921] Based on past purchase history, it predicts high demand during certain times of the day and suggests pricing for those times. The user checks the pricing, and the sentiment analysis engine analyzes their reactions, readjusts the price, and automatically sends it to the POS system.

[0922] 3. Automatic shift creation:

[0923] If a high number of customers are predicted on certain weekends, a shift schedule is generated that increases the number of employees required during those times. Users can check the schedule, and the sentiment analysis engine analyzes reactions, adjusts it as necessary, and distributes it to employees.

[0924] 4. Service and product demand forecast:

[0925] Based on past data, certain products are predicted to be popular in the next month and are listed as promotional items. Users check the list, and a sentiment analysis engine analyzes reactions and makes adjustments, planning promotions, and replenishing inventory.

[0926] Example prompt sentence:

[0927] 1. "Based on the sales data from the past year, forecast next week's demand and calculate the optimal order quantity."

[0928] 2. "Based on past customer purchasing history, identify times of day when demand is high and suggest pricing for those times."

[0929] 3. "Generate optimal shift schedules based on employee shift history and store visit predictions."

[0930] 4. "Analyze sales data and create a recommended list of products to promote and replenish inventory for next month."

[0931] This completes the description of the embodiment of the invention. This system enables integrated data management and efficient business operations, and realizes highly accurate predictions and optimization that reflect user feedback.

[0932] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0933] Step 1: Data collection

[0934] The server retrieves customer purchasing information, inventory data, disposal data, and employee shift information from the database. SQL queries are used as input, and raw data extracted from the database is obtained as output. Specifically, the server executes periodically scheduled jobs to retrieve the required data.

[0935] Step 2: Data Preprocessing

[0936] The server preprocesses the acquired raw data. The input contains the raw data extracted in step 1, and the output is the preprocessed clean data. Specific operations include imputing missing values ​​(for example, median imputation) and correcting outliers using the pandas library. Outlier detection uses common statistical methods (for example, z-score).

[0937] Step 3: Demand forecast

[0938] The server uses a generative AI model (for example, a time series analysis model using TensorFlow) based on the preprocessed data to make a demand forecast. The input is the data preprocessed in step 2, and the output is the demand forecast result. Specifically, past purchase data is input into the model to predict future demand. In this process, the time series data is analyzed using a model such as LSTM.

[0939] Step 4: Optimize order quantity

[0940] The server optimizes the order quantity based on the demand forecast results. The input is the demand forecast results obtained in step 3, and the output is an optimized order list. Specifically, the demand forecast results are input into an optimization algorithm (e.g., linear programming) to calculate the order quantity.

[0941] Step 5: Optimize pricing

[0942] The server optimizes pricing based on the demand forecast results. The input is the demand forecast results obtained in step 3, and the output is an optimized price list. Specifically, it uses an AI model (e.g., regression analysis) to analyze the relationship between demand and price and sets prices.

[0943] Step 6: Automatic shift creation

[0944] The server automatically generates employee shifts based on the demand forecast results. The inputs are the demand forecast results obtained in step 3 and employee desired shift information, and the output is an optimal shift schedule. Specifically, the optimal shifts are generated using a shift generation algorithm (e.g., a Bayesian network).

[0945] Step 7: User confirmation and feedback

[0946] The user checks the generated order list, price list, and shift schedule. The input is the data provided by the server, and the output is the user's feedback. In concrete terms, the user checks the data on the web application, and the sentiment analysis engine analyzes the user's feedback. For example, the user may enter text feedback such as "This order quantity is too large."

[0947] Step 8: Analyze feedback and adjust

[0948] The server uses an emotion analysis engine to analyze user feedback and adjust prediction and optimization results as necessary. The input is user feedback, and the output is adjusted order lists, price lists, and shift schedules. Specifically, the emotion analysis engine analyzes the user's text and facial expressions and adjusts the data to reflect the results.

[0949] Step 9: Autorun and Notifications

[0950] The server automatically takes the necessary actions based on the adjusted information and notifies the relevant parties. The input is the data adjusted in step 8, and the output is the sent order list, updated price information, and distributed shift schedules. Specific operations include sending order information to suppliers using APIs, updating price information in the POS system, and distributing shift information to employees.

[0951] (Application example 2)

[0952] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0953] Conventional business operation systems have difficulty in efficiently optimizing demand forecasting, inventory management, pricing, and employee shift scheduling, and also have problems in that they do not respond to the burden and emotions felt by users. For these reasons, there is a need to efficiently operate business operations in physical stores and improve user satisfaction.

[0954] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0955] In this invention, the server includes means for collecting customer purchasing information, inventory data, disposal data, and employee shift information, means for preprocessing the collected data and complementing missing and outlier values, means for using an AI model based on the preprocessed data to perform demand forecasting, inventory order optimization, pricing optimization, and automatic employee shift generation, means for presenting necessary information to a user based on the prediction results and generated data and for confirmation and correction, means for automatically performing necessary actions based on the information confirmed and corrected by the user, means for collecting emotion data and adjusting information based on the user's emotion, and means for providing and managing information in real time via mobile terminals and wearable devices, thereby enabling smart, emotion-conscious business operations.

[0956] "Customer purchasing information" refers to data such as information about the products purchased by the customer, the date and time of purchase, and frequency of purchase.

[0957] "Inventory data" refers to data that indicates the current inventory status of products in stores and warehouses.

[0958] "Disposal data" refers to data relating to the disposal status of a product, the reason for disposal, and the date and time of disposal.

[0959] "Employee shift information" refers to information such as the work schedules and desired holidays of store employees.

[0960] "Preprocessing" is the process of complementing missing or outliers in collected data and preparing it in a form suitable for input into an AI model.

[0961] An "AI model" is an algorithm that uses machine learning and deep learning to predict demand and optimize data.

[0962] "Demand forecasting" is the process of predicting how much and how little a customer will purchase of a particular product or service.

[0963] "Optimizing inventory order quantities" means determining the appropriate order quantity, neither too much nor too little, based on demand forecasts.

[0964] "Pricing optimization" means setting the optimal selling price for a product, taking into account supply and demand conditions.

[0965] "Automatic generation of employee shifts" means automatically creating optimal shift schedules by taking into account store visitor forecasts and employees' desired holidays.

[0966] "Emotion data" is data based on the recognition of the user's reactions and emotions.

[0967] "Mobile terminals and wearable devices" refers to portable devices such as smartphones, smart glasses, and head-mounted displays.

[0968] "Means for providing and managing information in real time" refers to a method for obtaining data instantly, providing information to users, and efficiently managing business operations.

[0969] This invention provides a system for optimizing the business operations of a physical store based on customer purchase information, inventory data, disposal data, and employee shift information. The operation of the system is described in detail below.

[0970] System configuration

[0971] This system mainly consists of four main components: a server, a mobile terminal, a wearable device, and a user. The roles of each are as follows:

[0972] Server Roles

[0973] The server comprehensively manages customer purchasing information, inventory data, disposal data, and employee shift information. Specifically, it collects and preprocesses this data and applies AI models to perform the following tasks:

[0974] Data collection and preprocessing

[0975] The server automatically collects various data and performs preprocessing to fill in missing and outlier values. Specific software used is the Python pandas library and scikit-learn.

[0976] Demand Forecasting and Optimization

[0977] Based on the preprocessed data, an AI model (for example, a deep learning model using TensorFlow or Keras) is used to predict demand for the following week and optimize order quantities and pricing.

[0978] Emotion data capture

[0979] It collects user emotional data and adjusts prediction results based on this data. The emotion engine uses an NLP model to analyze emotions from user text and voice input.

[0980] Autorun

[0981] Based on the information confirmed and corrected by the user, necessary actions (such as sending an order list or updating a shift schedule) are automatically performed.

[0982] Mobile and wearable devices

[0983] Mobile devices (such as smartphones) and wearable devices (such as smart glasses and head-mounted displays) serve as tools to provide users with information in real time and assist in management.

[0984] Real-time information provision

[0985] Inventory information, sales status, employee shift information, and more are instantly presented to users via mobile devices or wearable devices.

[0986] User Interface

[0987] These devices allow for voice and touch input as user interfaces, allowing users to easily check and modify information.

[0988] User operations

[0989] Users can check the information provided in real time and adjust order lists, pricing, and shift schedules as needed. These adjustments are fed back to the emotion engine and reflected in optimization.

[0990] Component Examples

[0991] A concrete example is an inventory management assistant for a brick-and-mortar store, where store staff can use smart glasses to check inventory information in real time, view items that need to be ordered, and forecast demand for specific times of day. The assistant also uses an emotion engine to collect staff emotions and automatically adjust forecasts accordingly.

[0992] Prompt Sentence Examples

[0993] To manage your store's inventory in real time, please provide us with a dataset that includes the following items:

[0994] Product ID

[0995] Product name

[0996] Current Stock

[0997] Inventory quantity one week ago

[0998] Disposal Data

[0999] Past sales data

[1000] In this way, collecting emotional data and optimizing business operations will enable efficient business operations in physical stores.

[1001] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1002] Step 1:

[1003] The server collects customer purchasing information, inventory data, waste data, and employee shift information through APIs. To obtain this data, the server uses HTTP requests to retrieve the necessary information from databases and external services. The input data is in JSON or CSV format, and the output is the collected raw data.

[1004] Step 2:

[1005] The server preprocesses the collected raw data. Specifically, it converts it into a data frame using the pandas library and imputes missing or outlier values. It also normalizes and standardizes the data as needed. This results in data in a format suitable for input to the AI ​​model. The input is the collected raw data, and the output is the preprocessed data.

[1006] Step 3:

[1007] The server applies AI models to the preprocessed data to forecast demand and optimize inventory order quantities. Specifically, it loads trained models using TensorFlow and Keras and executes predictions. The input is the preprocessed data, and the output is the prediction results.

[1008] Step 4:

[1009] The server collects the user's emotional data. The emotional data is obtained by analyzing the user's text and voice input with an NLP model. Specifically, it uses Hugging Face's Transformers for emotion analysis. The input is the user's text and voice data, and the output is the analyzed emotional data.

[1010] Step 5:

[1011] The server adjusts the prediction results based on the analyzed emotional data. By incorporating emotional data into the prediction results, more realistic and appropriate results can be obtained. For example, if there is a lot of negative emotional feedback, the demand forecast volume can be reduced. The inputs are the prediction results and emotional data, and the output is the adjusted prediction results.

[1012] Step 6:

[1013] The mobile terminal or wearable device provides the adjusted prediction results to the user in real time. The user can check the information through the device and make corrections as needed. Corrections are made via voice or touch input, and the data is sent back to the server. The input is the adjusted prediction results, and the output is the corrected prediction results or user feedback.

[1014] Step 7:

[1015] The server automatically executes the final action based on the user-modified information, such as sending the final inventory order list to suppliers or notifying employees of shift schedules. The input is the modified prediction results and feedback, and the output is the specific action to be taken.

[1016] Through the above steps, the server, mobile terminal, wearable device, and user can work together to optimize the business operations of a physical store and improve user satisfaction.

[1017] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1018] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1019] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[1020] [Third embodiment]

[1021] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1022] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[1023] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1024] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[1025] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1026] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1027] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1028] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1029] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1030] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1031] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1032] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[1033] This invention relates to a system for optimizing specific business operations, and aims to improve business efficiency by applying AI models to mainly utilize customer purchasing information, inventory data, waste data, and employee shift information. The system consists of three main components: a server, a terminal, and a user.

[1034] Automated ordering system

[1035] Server: In the ordering system, the server first collects inventory status, past order data, and waste data. Next, the server preprocesses this data and fills in missing and outlier values. Based on the preprocessed data, the server uses an AI model to forecast demand and calculate the next order quantity.

[1036] For example, if an analysis of data from the past year predicts that demand for a particular product will increase next week, the server automatically generates an appropriate order quantity based on that prediction. After the user checks this order list and makes any necessary corrections, the server automatically sends the final list to the supplier.

[1037] Dynamic Pricing System

[1038] Server: In this system, the server collects customer purchase history and store visits by time of day, and pre-processes this data. Based on the pre-processed data, the server uses an AI model to analyze the relationship between demand and price. Next, it sets the optimal price based on demand and automatically distributes it to the POS system.

[1039] For example, if past purchase history shows that demand is highest between 3:00 PM and 6:00 PM on weekdays, you can set the price to be 10% higher during that time period, which will reduce unsold items and maximize profits.

[1040] Automatic shift creation

[1041] Server: In the shift creation system, the server collects employee shift history, desired vacation time, and store visit predictions. It then preprocesses this data and uses an AI model to predict optimal shifts based on the number of employees and peak hours. The predicted shift schedule is automatically generated and can be reviewed and modified by the user.

[1042] For example, if a high number of customers is predicted on a particular weekend, the system generates a shift schedule that automatically allocates the necessary number of employees during that time period. The user can check the generated schedule, correct any problems, and automatically distribute it to employees.

[1043] Service and product demand forecasting

[1044] Server: In a demand forecasting system, the server collects and preprocesses customer purchase history and sales data. Based on the preprocessed data, the server uses an AI model to predict future demand. Based on the prediction results, it generates a list of recommended promotional items and inventory replenishment and presents it to the user.

[1045] For example, if past data predicts that a particular product will be popular in the next month, the product will be listed as a promotion target. Users can check the list and make appropriate promotion plans and inventory replenishment.

[1046] In this way, the invention will reduce food waste, alleviate labor shortages, and achieve sustainable service provision through appropriate pricing.

[1047] The processing flow will be explained below.

[1048] Automated ordering system processing

[1049] Step 1:

[1050] Server: Collects inventory status, past order data, and disposal data from the database.

[1051] Every day at 2:00 AM, the server collects inventory status, order data, and disposal data from the database for the past year.

[1052] Step 2:

[1053] Server: Preprocesses the collected data and imputes missing or outlier values.

[1054] After collecting the data, the server performs preprocessing, imputing missing values ​​with the average value and removing outliers.

[1055] Step 3:

[1056] Server: Preprocessed data is input into the AI ​​model to generate demand forecasts.

[1057] The server passes the pre-processed data to an AI model to generate demand forecasts for the next week.

[1058] Step 4:

[1059] Server: Calculates the optimal order quantity based on the demand forecast results.

[1060] The server calculates the optimal order quantity for each product based on the demand forecast results output by the AI ​​model.

[1061] Step 5:

[1062] Server: Automatically generate an order list based on the optimal order quantity.

[1063] The server creates an order list based on the calculated order quantity.

[1064] Step 6:

[1065] User: Check the generated order list and modify it as necessary.

[1066] The user checks the order list on the system and corrects the quantity if necessary.

[1067] Step 7:

[1068] Server: Automatically sends the confirmed and corrected order list to the supplier.

[1069] The server receives the order list that the user has confirmed and modified, and automatically sends it to the specified supplier.

[1070] Dynamic pricing system processing

[1071] Step 1:

[1072] Server: Collects customer purchase history and store visit data by time of day.

[1073] Every night, the server collects customer purchase history and time-of-day visit data from the database.

[1074] Step 2:

[1075] Server: Preprocesses the collected data and removes outliers.

[1076] The server performs preprocessing after data collection to normalize or remove outliers.

[1077] Step 3:

[1078] Server: Inputs the preprocessed data into the AI ​​model and analyzes the relationship between demand and price.

[1079] The server passes the pre-processed data to an AI model that analyzes the relationship between demand and price.

[1080] Step 4:

[1081] Server: Sets the price for the next day based on the analysis results.

[1082] The server uses the analysis results output by the AI ​​model to set the optimal price for the next day.

[1083] Step 5:

[1084] Server: Automatically distributes the set price information to the POS system.

[1085] The server sends the set price information to the POS system, which updates it automatically.

[1086] Process automatic shift creation

[1087] Step 1:

[1088] Server: Collects employee shift history, desired vacation time, and store visit forecast data.

[1089] At the end of each month, the server collects employee shift history, desired vacation data, and store visit forecast data from the database.

[1090] Step 2:

[1091] Server: Preprocesses the collected data and imputes missing or outlier values.

[1092] After data collection, the server performs preprocessing, imputing missing values ​​and removing outliers.

[1093] Step 3:

[1094] Server: Based on the pre-processed data, the AI ​​model predicts the optimal shift.

[1095] The server inputs the pre-processed data into an AI model to predict the optimal shift schedule.

[1096] Step 4:

[1097] Server: Automatically generates predicted shift schedules.

[1098] The server automatically generates shift schedules based on the prediction results output by the AI ​​model.

[1099] Step 5:

[1100] User: Review the generated shift schedule and modify it if necessary.

[1101] Users can check the shift schedule on the system and modify it if necessary.

[1102] Step 6:

[1103] Server: Automatically distributes confirmed shift schedules to employees.

[1104] The server receives the shift schedule that the user has confirmed and modified and automatically distributes it to all employees.

[1105] Processing demand forecasts for services and products

[1106] Step 1:

[1107] Server: Collects customer purchase history and sales data.

[1108] The server collects customer purchase history and sales data from the database every day at 11:00 p.m.

[1109] Step 2:

[1110] Server: Preprocesses the collected data and imputes missing or outlier values.

[1111] After data collection, the server performs preprocessing, imputing missing values ​​and removing outliers.

[1112] Step 3:

[1113] Server: Inputs the pre-processed data into the AI ​​model to predict future demand.

[1114] The server passes the pre-processed data to an AI model to predict future demand.

[1115] Step 4:

[1116] Server: Based on the prediction results, a list of recommended promotion targets and inventory replenishment is generated.

[1117] The server automatically generates a list of recommended promotional targets and inventory replenishment based on the prediction results output by the AI ​​model.

[1118] Step 5:

[1119] User: Review the generated recommendation list and incorporate it into promotion planning and inventory replenishment.

[1120] The user checks the recommended list and implements promotion plans and inventory replenishment as necessary.

[1121] Example 1

[1122] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1123] In conventional business systems, customer purchasing information, inventory data, disposal data, and employee shift information are often managed separately, making it difficult to integrate and efficiently utilize this data. Furthermore, business processes such as demand forecasting, pricing, and automatic shift generation are performed manually, requiring a great deal of time and effort, making it difficult to improve the efficiency and accuracy of business operations. Aiming to solve these problems, the present invention provides a system for optimizing specific business operations.

[1124] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1125] In this invention, the server includes means for collecting customer purchasing information, inventory data, disposal data, and employee shift information, means for preprocessing the collected data and complementing missing values ​​and outliers, means for using a generative AI model based on the preprocessed data to perform demand forecasting, optimizing inventory order quantities, optimizing pricing, and automatically generating employee shifts, means for presenting necessary information to a user based on the prediction results and generated data and for confirmation and correction, and means for automatically executing necessary actions based on the information confirmed and corrected by the user. This enables efficient management of customer purchasing information, inventory data, disposal data, and employee shift information, automating business processes such as demand forecasting, inventory ordering, pricing, and shift creation, and improving business efficiency and accuracy.

[1126] Key Word Definitions

[1127] "Customer purchasing information" refers to data that records the type, quantity, purchase date, etc. of products that a customer has purchased in the past.

[1128] "Inventory data" refers to data for managing the quantity of products currently in stock, storage locations, and inventory history.

[1129] "Disposal data" refers to data that records the quantity of products that have been discarded due to reasons such as expiration or poor quality, the reason for disposal, and the date and time of disposal.

[1130] "Employee shift information" refers to data necessary for labor management, such as employees' working hours, desired holidays, and shift history.

[1131] "Missing values" refers to a state in which data that should be present in a dataset is missing.

[1132] An "outlier" is a value that is significantly different from other data in a data set and is typically treated as an error or outlier.

[1133] A "generative AI model" is a model that uses artificial intelligence technology to learn specific patterns from data and automatically perform tasks such as prediction and classification.

[1134] "Demand forecasting" is the process of predicting future demand based on past data, and refers to predicting sales quantities, usage amounts, etc.

[1135] "Inventory order optimization" is the process of determining optimal inventory levels based on demand forecasts to avoid shortages and excess inventory.

[1136] "Pricing optimization" is the process of setting optimal selling prices taking into account demand and market conditions, with the goal of maximizing profits.

[1137] "Automatic employee shift generation" is the process of automatically creating employee work schedules, taking into account business demands and employee preferences.

[1138] "Means for automatically taking necessary actions based on information confirmed and corrected by the user" refers to a process for automatically taking business actions such as placing orders, changing prices, and notifying shifts based on data confirmed and corrected by the user.

[1139] MODE FOR CARRYING OUT THE INVENTION

[1140] This invention relates to a system for optimizing specific business operations, and aims to improve business efficiency by applying generative AI models primarily using customer purchasing information, inventory data, waste data, and employee shift information. This system consists of three main components: a server, a terminal, and a user.

[1141] Hardware and software used

[1142] The server is built on AWS (a common cloud computing service) and uses MySQL as the database. Python and libraries such as Pandas and Scikit-learn are used for data preprocessing and analysis. AI frameworks such as TensorFlow, Keras, XGBoost, and Prophet are mainly used to implement the generative AI model. A general sales management system is assumed as the POS system.

[1143] System example and operation

[1144] Automated ordering system

[1145] Server: The server periodically collects inventory status, past order data, and waste data from terminals and point-of-sale (POS) systems, and preprocesses this data using Python and Pandas. It imputes missing and outlier values ​​and inputs the data into a generative AI model (e.g., TensorFlow) to perform demand forecasting. It automatically generates an order list based on the forecast results and presents it to the user. The user can review and modify the order list, and the final list is automatically sent from the server to the supplier.

[1146] Example: Based on data from the past year, the order quantity for the next week, when demand is predicted to increase, is automatically calculated and an order list is created.

[1147] Dynamic Pricing System

[1148] Server: The server collects and preprocesses customer purchase history and store visit status by time of day in real time. This data is input into a generative AI model (e.g., XGBoost) to analyze the relationship between demand and price. The optimal price is calculated from the analysis results and automatically distributed to the sales management system.

[1149] Example: If analysis shows that demand is high between 3pm and 6pm on weekdays, the price during that time period will be increased by 10%.

[1150] Automatic shift creation

[1151] Server: The server collects and preprocesses employee shift history, desired vacation time, and store visit prediction information. This data is input into a generative AI model (e.g., Keras) to predict and automatically generate employee shift schedules. Users can check and modify the generated shift schedules. The final shift schedules are automatically distributed from the server to employees.

[1152] Example: If store visits are predicted to be high on a particular weekend, create a shift that places more employees at that time.

[1153] Service and product demand forecasting

[1154] Server: The server periodically collects customer purchase history and sales data, and after preprocessing, inputs it into a generative AI model (e.g., Prophet) to predict future demand. Based on the prediction results, it generates a list of recommended promotional items and inventory replenishment and presents it to the user. The user checks this list and takes the necessary actions.

[1155] Example: If historical data predicts that a particular product will be popular next month, target that product for promotion.

[1156] Prompt Sentence Examples

[1157] "Please forecast demand for the next month based on data from the past year, and propose periods when demand will be high and product promotions. Please also let us know specific promotion ideas."

[1158] In this way, the present invention is a system that uses a generative AI model to integrate and analyze a variety of business data and automatically implement optimized business processes, thereby improving business efficiency and supporting highly accurate decision-making.

[1159] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1160] Processing steps of specific business operation optimization system

[1161] Automated ordering system processing flow

[1162] Step 1: Data collection

[1163] Server: The server periodically collects inventory status, past order data, and waste data from terminals and POS systems. Raw data obtained from each terminal is sent to the server as input. The output is consolidated data stored in a database within the server. As a concrete example, an automatic data collection script is executed every night.

[1164] Step 2: Data Preprocessing

[1165] Server: The collected data is preprocessed using Python and Pandas. The collected raw data is used as input. Missing values ​​and outliers are imputed, and the output is formatted as time series data. For example, missing values ​​are imputed with the past average value, and outliers with the median value.

[1166] Step 3: Demand forecasting using AI models

[1167] Server: The preprocessed data is input into a generative AI model (e.g., TensorFlow) to generate a demand forecast. The forecast model takes into account trends and seasonal fluctuations based on past data. The preprocessed data is used as input, and a forecast of future demand is output. For example, if demand for next week is predicted to be 200 units, the server outputs that number.

[1168] Step 4: Generate an order list

[1169] Server: Automatically generates an order list based on the forecast results. Based on the forecasted demand, the optimal order quantity is calculated taking into account the safety stock amount, and compiled in list format. The demand forecast results are used as input, and the output is an order list. As a specific example, if the forecasted demand is 200 units, an order list for 220 units is generated taking into account the safety stock amount.

[1170] Step 5: Check and modify the order list

[1171] User: The user checks the generated order list in a dedicated interface and modifies it if necessary. The automatically generated order list is used as input, and a modified final list is generated as output. For example, if the user feels that there are too many of a particular product in this list, the order quantity can be adjusted by the user.

[1172] Step 6: Send the order list to your suppliers

[1173] Server: Automatically sends the final purchase order list, which has been reviewed and modified by the user, to the supplier. It uses the modified purchase order list as input and generates purchase instructions as output, which are sent to the supplier, for example, via email or electronic data interchange (EDI).

[1174] Dynamic pricing system processing flow

[1175] Step 1: Data collection

[1176] Server: The server collects customer purchase history and store visit status by time of day in real time. The input is raw data obtained from the POS system, and the output is purchase data stored in a database within the server. For example, the latest data is obtained from the POS system every hour.

[1177] Step 2: Data Preprocessing

[1178] Server: Cleanses the collected data and removes outliers using Python and Scikit-learn. The input is raw data collected from the POS system, and the output is a clean dataset. For example, the format of dates and times is standardized.

[1179] Step 3: Analyze the relationship between demand and price

[1180] Server: The preprocessed data is input into an AI model (e.g., XGBoost) to analyze the relationship between demand and price. The input is the preprocessed data, and the output is the analysis result. For example, demand forecasting is performed using the relationship between past price fluctuations and demand.

[1181] Step 4: Set the optimal price

[1182] Server: Sets the optimal price based on the analysis results. The analysis results are used as input and new pricing is generated as output. For example, the price is set to increase by 10% during times of high demand.

[1183] Step 5: Distributing price information to POS systems

[1184] Server: Calculated price information is sent to the POS system in real time, automatically updating the store's prices. The input is the new price setting, and the output is the price information reflected in the POS system. For example, the new price information is immediately applied to the POS system and displayed to the customer.

[1185] Process flow for automatic shift creation

[1186] Step 1: Data collection

[1187] Server: The server collects employee shift history, desired vacation time, and store visit prediction information. The employee database and store visit prediction data are used as input, and the output is the integrated data stored in the server. For example, the latest data is obtained every Sunday.

[1188] Step 2: Data Preprocessing

[1189] Server: Cleanses the collected data and processes missing and outliers using Python and Pandas. The input is raw data, and the output is a clean dataset. For example, missing values ​​are imputed with the mean and outliers are removed.

[1190] Step 3: Shift prediction using AI models

[1191] Server: Input the preprocessed data into an AI model (e.g., Keras) to predict employee shifts. The input is the preprocessed data, and the output is the predicted shift schedule. For example, it allocates the required number of employees according to peak hours.

[1192] Step 4: Generate a shift schedule

[1193] Server: Automatically generates shift schedules based on the prediction results. The input is the prediction results, and the output is the generated shift schedule. As a specific example, shifts are adjusted according to the predicted number of customers.

[1194] Step 5: Check and modify the shift schedule

[1195] User: The user checks the generated shift schedule and modifies it if necessary. The input is the automatically generated shift schedule, and the output is the final modified schedule. For example, if a user requests sudden leave, the shift can be covered by another employee.

[1196] Step 6: Distribute shift schedules to employees

[1197] Server: Automatically distributes the final shift schedule to employees. The input is the revised shift schedule, and the output is the shift information sent to employees. For example, it can be sent via email or a dedicated app.

[1198] Service / product demand forecasting process flow

[1199] Step 1: Data collection

[1200] Server: The server periodically collects customer purchase history and sales data. The input is raw data obtained from the POS system, and the output is sales data stored in a database on the server. For example, data is collected at the end of each month.

[1201] Step 2: Data Preprocessing

[1202] Server: Cleanses the collected data and processes missing values ​​and outliers using Python and Pandas. The input is raw data, and the output is a clean dataset. For example, missing values ​​are imputed with the average value of the previous month, and outliers are removed.

[1203] Step 3: Demand forecasting using AI models

[1204] Server: Inputs the preprocessed data into a generative AI model (e.g., Prophet) to predict future demand. The input is the preprocessed data, and the output is the prediction result. For example, a specific product is expected to be in high demand next month.

[1205] Step 4: Generate a recommendation list

[1206] Server: Based on the prediction results, a list of recommendations for promotions and inventory replenishment is generated. The input is the prediction results, and the output is the generated recommendation list. As a concrete example, it lists products with increasing demand.

[1207] Step 5: Review and modify the recommendation list

[1208] User: The user reviews the generated recommendation list and modifies it if necessary. The input is the automatically generated recommendation list, and the output is the modified list. For example, the user determines that a critical item needs to be restocked.

[1209] Step 6: Planning promotions and replenishing inventory

[1210] User: After reviewing the recommendation list, create promotion plans or replenish inventory based on the list. The input is the revised recommendation list, and the output is the implemented promotions or replenished inventory. For example, target specific products for a special campaign.

[1211] (Application example 1)

[1212] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1213] In traditional brick-and-mortar stores, inventory management, pricing, demand forecasting, and employee shift planning are often done manually, resulting in problems of inefficiency and human error in each process. Furthermore, there was a lack of means to grasp trends in customer purchasing behavior in real time and respond quickly, making appropriate inventory management and pricing difficult. Furthermore, it was difficult to implement automatic ordering based on forecasts or dynamic pricing to prevent unsold items, making it difficult to develop effective sales strategies. It is essential to solve these problems and improve the overall efficiency of store operations.

[1214] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1215] In this invention, the server includes: means for collecting customer purchasing information, inventory data, disposal data, and employee shift information; means for preprocessing the collected data and complementing missing and outlier values; means for using an AI model based on the preprocessed data to forecast demand, optimize inventory order quantities, optimize pricing, and automatically generate employee shifts; means for presenting necessary information to a user based on the forecast results and generated data and allowing the user to confirm and correct it; means for automatically executing necessary actions based on the information confirmed and corrected by the user; means for providing inventory management functions in real time and issuing alerts to prevent stockouts and excess inventory; means for dynamically setting product prices based on purchase history and the number of customers by time period to prevent unsold items and maximize profits; means for automatically creating optimal shift schedules based on employee shift history and desired vacation days; and means for predicting future demand based on customer purchase history and sales data and generating recommended lists of promoted products and inventory replenishment. This allows for centralized management of various processes related to store operations, enabling efficient and accurate inventory management, pricing, demand forecasting, and shift creation.

[1216] "Customer purchasing information" is data including the type and quantity of products purchased by customers who visit a store, the date and time of purchase, and the attributes of the purchaser.

[1217] "Inventory data" refers to information relating to the current quantity of products held in the store, storage locations, replenishment status, and the like.

[1218] "Disposal data" is data indicating the type, quantity, and date and time of disposal of products that have been discarded due to the expiration of their sales period or deterioration of quality.

[1219] "Employee shift information" is data including an employee's working date and time, working hours, vacation days, desired work shifts, etc.

[1220] "Preprocessing" is the process of converting collected data into a format suitable for analysis and model training, and filling in missing or outlier values.

[1221] An "AI model" is a mathematical model that uses artificial intelligence technologies such as machine learning and deep learning to make useful predictions and classifications from data.

[1222] "Demand forecasting" is the process of predicting future demand for a product based on past data.

[1223] "Inventory order quantity optimization" is the process of preventing excess inventory and out-of-stock by predicting product demand and calculating the appropriate order quantity based on that.

[1224] "Pricing optimization" is the process of adjusting the price of a product to an optimal value based on demand data, etc.

[1225] "Automatic generation of employee shifts" is the process of automatically creating optimal shift schedules based on employees' shift history, desired working hours, and store business demands.

[1226] "Prediction results and generated data" refers to information such as demand forecast results, optimal order quantities, appropriate prices, and shift schedules generated by AI models.

[1227] The "inventory management function" monitors inventory status in real time and reduces the risk of stockouts and excess inventory.

[1228] "Dynamic setting" means that values ​​and parameters can be flexibly changed according to real-time data and the current situation.

[1229] "Promotional Products" means selected products that are specially promoted or discounted for promotional purposes.

[1230] A "recommended list" is a list generated by the system to encourage specific actions or responses.

[1231] This invention is built as a system for optimizing specific business operations. The system consists of three main components: a server, a terminal, and a user.

[1232] Inventory management function

[1233] The server first collects customer purchasing information, inventory data, waste data, and employee shift information. This data is obtained from databases and POS systems. The server then preprocesses the data and fills in missing and outlier values. Specifically, it cleanses the data using data analysis libraries such as Pandas. Using the preprocessed data, it trains a demand forecasting model using AI libraries such as TensorFlow and Keras. This model predicts future inventory demand and calculates appropriate order quantities.

[1234] Dynamic Pricing Feature

[1235] The server also analyzes purchase history and store visits by time of day to determine optimal pricing based on demand. This data is also preprocessed and fed into the AI ​​model, which analyzes the relationship between demand and price and calculates the most profitable price. This price information is automatically sent to the POS system, where prices are adjusted in real time.

[1236] Automatic shift creation

[1237] The server automatically generates optimal shift schedules based on employees' shift history, desired holidays, and predicted store visits. Employee information and store visit prediction data are fed into an AI model to calculate the optimal shift schedule. Users can check and modify the generated schedule, and the final shift schedule is distributed to employees.

[1238] Demand forecasting and promotions

[1239] The server also collects customer purchase history and sales data to predict future demand. Based on this prediction, it targets specific products for promotion and generates a list of recommended inventory replenishment. Specifically, if the AI ​​model predicts that a particular product will be popular in the next month, it will list that product as a promotion target.

[1240] Hardware and software used

[1241] The hardware used is a server for database hosting and AI model training, and a smartphone as a client terminal. The software used is Pandas (data collection and preprocessing) and TensorFlow / Keras (demand forecasting AI model construction and training).

[1242] Examples of specific examples and prompts

[1243] For example, if a particular product is predicted to be popular in the coming month, the product may be listed for promotion and the store may be notified.

[1244] Example prompt sentence:

[1245] "Using past sales data, the AI ​​model predicts which products will be in high demand for the next month, and creates a promotion plan for those products."

[1246] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1247] Step 1:

[1248] The server collects customer purchasing information, inventory data, disposal data, and employee shift information.

[1249] These data are obtained from databases and POS systems. The collected data is first converted into a data frame using the Pandas library, and missing and outlier values ​​are imputed (data cleansing).

[1250] Input: Customer purchasing information, inventory data, disposal data, employee shift information

[1251] Output: Preprocessed data frame

[1252] Step 2:

[1253] The server trains the AI ​​model based on the preprocessed data.

[1254] Build and train a demand forecasting model using TensorFlow and Keras libraries. Historical purchase and inventory data are used as input data for the model to generate demand forecasts.

[1255] Input: Preprocessed data frame

[1256] Output: A trained demand forecasting model

[1257] Step 3:

[1258] The server uses the trained AI model to predict future inventory demand and calculate appropriate order quantities.

[1259] Past purchasing data is input into the forecasting model to predict demand for the next week or month. Based on the forecast results, the optimal inventory order quantity is calculated.

[1260] Inputs: Trained demand forecasting model, historical purchase data

[1261] Output: Future inventory demand forecast, order quantity

[1262] Step 4:

[1263] The server automatically generates an order list based on the demand forecast results and the order quantity, and presents it to the user.

[1264] The generated order list is displayed on the terminal for the user to review and modify, and after the user makes the appropriate modifications, the information is automatically sent to the supplier.

[1265] Input: Future inventory demand forecast, order quantity

[1266] Output: Order list, order data to suppliers

[1267] Step 5:

[1268] The server analyzes purchase history and store visits by time of day, and sets optimal prices based on demand.

[1269] The AI ​​model analyzes the relationship between demand and price to calculate the optimal price, which is then automatically sent to the POS system.

[1270] Input: Purchase history, store visits by time period

[1271] Output: Best price, price information to POS system

[1272] Step 6:

[1273] The server automatically generates an optimal shift schedule based on employee shift history, desired holidays, and predicted store visits, and presents it to the user.

[1274] The AI ​​model calculates the optimal shift schedule, which the user can then review and modify. The modified schedule is then distributed to employees.

[1275] Input: Shift history, desired leave, store visit forecast

[1276] Output: Shift schedule, information distributed to employees

[1277] Step 7:

[1278] The server predicts future demand based on customer purchase history and sales data, generates a recommended list of promotional items and inventory replenishment, and presents it to the user.

[1279] Based on the generated recommendation list, the user plans promotions and stock replenishment.

[1280] Input: Purchase history, sales data

[1281] Output: Demand forecast, recommendation list, promotional products

[1282] The above are the specific processing steps for carrying out the present invention.

[1283] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1284] This invention relates to a system for optimizing specific business operations, primarily using customer purchasing information, inventory data, waste data, and employee shift information to apply AI models to improve business efficiency. Furthermore, by combining it with an emotion engine that recognizes user emotions, even more accurate predictions and optimization can be achieved. This system consists of three main components: a server, a terminal, and a user.

[1285] Automated ordering system

[1286] Server: In the ordering system, the server first collects inventory status, past order data, and waste data. Next, the server preprocesses this data and fills in missing and outlier values. Based on the preprocessed data, the server uses an AI model to forecast demand and calculate the next order quantity. It also uses an emotion engine to incorporate user feedback and adjust the optimal order quantity.

[1287] For example, if an analysis of data from the past year predicts that demand for a particular product will increase next week, the system automatically generates an appropriate order quantity based on that prediction. After the user reviews this order list and the emotion engine analyzes the user's reactions and makes any necessary adjustments, the server automatically sends the final list to the supplier.

[1288] Dynamic Pricing System

[1289] Server: In this system, the server collects customer purchase history and store visits by time of day, and pre-processes this data. Based on the pre-processed data, the server uses an AI model to analyze the relationship between demand and price. Next, the server sets the optimal price based on demand, and an emotion engine incorporates user feedback to adjust the price setting, which is then automatically distributed to the POS system.

[1290] For example, if past purchasing history shows that demand is high between 3:00 PM and 6:00 PM on weekdays, the price can be increased by 10% during that time period, reducing unsold items and maximizing profits. The emotion engine analyzes user reactions and readjusts prices based on feedback.

[1291] Automatic shift creation

[1292] Server: In the shift creation system, the server collects employee shift history, desired vacation time, and store visit predictions. It then preprocesses this data and uses an AI model to predict optimal shifts based on the number of employees and peak hours. The predicted shift schedule is automatically generated and confirmed by the user. An emotion engine also analyzes the user's reactions and adjusts the shift schedule as necessary.

[1293] As a specific example, if a high number of customers is predicted on a particular weekend, the system will automatically generate a shift schedule that allocates the necessary number of employees during that time period. The user can check the generated schedule, and the emotion engine will analyze the user's reactions, correct any problems, and automatically distribute the schedule to employees.

[1294] Service and product demand forecasting

[1295] Server: In a demand forecasting system, the server collects and preprocesses customer purchase history and sales data. Based on the preprocessed data, the server uses an AI model to predict future demand. Based on the prediction results, it generates a list of recommended promotional items and inventory replenishment and presents it to the user. An emotion engine analyzes user reactions and adjusts the recommended list.

[1296] For example, if past data predicts that a particular product will be popular in the next month, the product will be listed as a promotion target. Users can check the list, and the emotion engine will analyze their reactions and make appropriate adjustments, such as planning promotions and replenishing inventory.

[1297] In this way, the invention will realize sustainable service provision through food waste reduction, labor shortage mitigation, and appropriate pricing. In addition, by integrating an emotion engine, more accurate prediction and optimization will be possible, leading to improved user satisfaction.

[1298] The processing flow will be explained below.

[1299] Automated ordering system processing

[1300] Step 1:

[1301] Server: Collects inventory status, past order data, and disposal data from the database.

[1302] Every day at 2:00 AM, the server collects inventory status, order data, and disposal data from the database for the past year.

[1303] Step 2:

[1304] Server: Preprocesses the collected data and imputes missing or outlier values.

[1305] After collecting the data, the server performs preprocessing, imputing missing values ​​with the average value and removing outliers.

[1306] Step 3:

[1307] Server: Preprocessed data is input into the AI ​​model to generate demand forecasts.

[1308] The server passes the pre-processed data to an AI model to generate demand forecasts for the next week.

[1309] Step 4:

[1310] Server: Calculates the optimal order quantity based on the demand forecast results.

[1311] The server calculates the optimal order quantity for each product based on the demand forecast results output by the AI ​​model.

[1312] Step 5:

[1313] Server: Automatically generate an order list based on the optimal order quantity.

[1314] The server creates an order list based on the calculated order quantity.

[1315] Step 6:

[1316] User: Check the generated order list and modify it as necessary.

[1317] The user checks the order list on the system and corrects the quantity if necessary.

[1318] Step 7:

[1319] Emotion Engine: Analyzes user reactions and suggests adjustments to the order list.

[1320] The emotion engine analyzes user reactions in real time and suggests adjustments to order quantities.

[1321] Step 8:

[1322] Server: Automatically sends the confirmed and corrected order list to the supplier.

[1323] The server receives the order list that the user has confirmed and modified, and automatically sends it to the specified supplier.

[1324] Dynamic pricing system processing

[1325] Step 1:

[1326] Server: Collects customer purchase history and store visit data by time of day.

[1327] Every night, the server collects customer purchase history and time-of-day visit data from the database.

[1328] Step 2:

[1329] Server: Preprocesses the collected data and removes outliers.

[1330] The server performs preprocessing after data collection to normalize or remove outliers.

[1331] Step 3:

[1332] Server: Inputs the preprocessed data into the AI ​​model and analyzes the relationship between demand and price.

[1333] The server passes the pre-processed data to an AI model that analyzes the relationship between demand and price.

[1334] Step 4:

[1335] Server: Sets the price for the next day based on the analysis results.

[1336] The server uses the analysis results output by the AI ​​model to set the optimal price for the next day.

[1337] Step 5:

[1338] Emotion Engine: Analyzes user reactions and suggests pricing adjustments.

[1339] The sentiment engine analyzes user reactions in real time and suggests pricing adjustments.

[1340] Step 6:

[1341] Server: Automatically distributes the set price information to the POS system.

[1342] The server sends the set price information to the POS system, which updates it automatically.

[1343] Process automatic shift creation

[1344] Step 1:

[1345] Server: Collects employee shift history, desired vacation time, and store visit forecast data.

[1346] At the end of each month, the server collects employee shift history, desired vacation data, and store visit forecast data from the database.

[1347] Step 2:

[1348] Server: Preprocesses the collected data and imputes missing or outlier values.

[1349] After data collection, the server performs preprocessing, imputing missing values ​​and removing outliers.

[1350] Step 3:

[1351] Server: Based on the pre-processed data, the AI ​​model predicts the optimal shift.

[1352] The server inputs the pre-processed data into an AI model to predict the optimal shift schedule.

[1353] Step 4:

[1354] Server: Automatically generates predicted shift schedules.

[1355] The server automatically generates shift schedules based on the prediction results output by the AI ​​model.

[1356] Step 5:

[1357] User: Review the generated shift schedule and modify it if necessary.

[1358] Users can check the shift schedule on the system and modify it if necessary.

[1359] Step 6:

[1360] Emotion engine: Analyzes user reactions and suggests shift schedule adjustments.

[1361] The emotion engine analyzes user reactions in real time and suggests adjustments to shift schedules.

[1362] Step 7:

[1363] Server: Automatically distributes confirmed shift schedules to employees.

[1364] The server receives the shift schedule that the user has confirmed and modified and automatically distributes it to all employees.

[1365] Processing demand forecasts for services and products

[1366] Step 1:

[1367] Server: Collects customer purchase history and sales data.

[1368] The server collects customer purchase history and sales data from the database every day at 11:00 p.m.

[1369] Step 2:

[1370] Server: Preprocesses the collected data and imputes missing or outlier values.

[1371] After data collection, the server performs preprocessing, imputing missing values ​​and removing outliers.

[1372] Step 3:

[1373] Server: Inputs the pre-processed data into the AI ​​model to predict future demand.

[1374] The server passes the pre-processed data to an AI model to predict future demand.

[1375] Step 4:

[1376] Server: Based on the prediction results, a list of recommended promotion targets and inventory replenishment is generated.

[1377] The server automatically generates a list of recommended promotional targets and inventory replenishment based on the prediction results output by the AI ​​model.

[1378] Step 5:

[1379] User: Review the generated recommendation list and incorporate it into promotion planning and inventory replenishment.

[1380] The user checks the recommended list and implements promotion plans and inventory replenishment as necessary.

[1381] Step 6:

[1382] Emotion engine: Analyzes user reactions and suggests adjustments to the recommendation list.

[1383] The emotion engine analyzes user reactions in real time and makes suggestions to adjust the recommendation list.

[1384] In this way, the invention will realize sustainable service provision through food waste reduction, labor shortage mitigation, and appropriate pricing. In addition, by integrating an emotion engine, more accurate prediction and optimization will be possible, leading to improved user satisfaction.

[1385] Example 2

[1386] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1387] In conventional business operation systems, customer purchasing information, inventory data, disposal data, and employee shift information are managed separately, making data integration and management cumbersome and making efficient operation difficult. Furthermore, predictions and optimization based on this data are often performed manually, limiting the efficiency of operations. Furthermore, because feedback that takes user emotions into account is not reflected, optimization results may not meet on-site needs. Therefore, a system that can solve these issues all at once is needed.

[1388] The identification processing by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting customer purchasing information, inventory data, disposal data, and employee shift information, means for preprocessing the collected data and complementing missing values ​​and outliers, means for performing demand forecasting using a generative AI model based on the preprocessed data, means for optimizing order quantities, optimizing pricing, and automatically generating employee shifts based on the generated demand forecast, means for obtaining user feedback using a sentiment analysis engine based on the prediction results and the generated data and adjusting necessary information, and means for automatically executing necessary actions based on information confirmed and corrected by the user. This enables integrated data management and efficient business operations, and realizes highly accurate predictions and optimization that reflect user feedback.

[1389] "Customer purchasing information" refers to data such as the products purchased by the customer, the purchase date and time, and the quantity purchased.

[1390] "Inventory data" refers to data including information such as product inventory quantity, date of receipt, and storage location.

[1391] "Disposal data" refers to data that includes information such as the quantity of inventory items to be discarded, the reason for disposal, and the date and time of disposal.

[1392] "Employee shift information" refers to data such as employees' working days, working hours, and desired holidays.

[1393] "Means of collection" refers to the methods and equipment used to obtain the necessary information from databases and sensors.

[1394] "Preprocessing means" refers to a method for completing missing data values, eliminating outliers, and converting the data into a format suitable for analysis.

[1395] A "generative AI model" refers to a model that includes algorithms for demand forecasting and optimization from data using techniques such as machine learning and deep learning.

[1396] "Means for demand forecasting" refers to methods and systems that use AI models to estimate future demand.

[1397] "Order quantity optimization" refers to the process of determining the optimal product order quantity based on demand forecasts.

[1398] "Pricing optimization" refers to the process of determining the most effective price by taking into account the balance between supply and demand.

[1399] "Automatic employee shift generation" refers to a system that automatically creates work schedules based on employee preferences and business needs.

[1400] An "emotion analysis engine" refers to a technology that analyzes emotions from user input, facial expressions, and voice, and reflects that feedback in the system.

[1401] "Means for adjusting required information" refers to a method or system that adjusts predictions or optimization results based on user feedback obtained using a sentiment analysis engine.

[1402] "Means for automatically executing necessary actions" refers to a method or system whereby the system automatically executes the relevant business process based on the information confirmed or corrected by the user.

[1403] This invention relates to a system for optimizing specific business operations, which utilizes customer purchasing information, inventory data, waste data, and employee shift information to apply a generative AI model to improve business efficiency. Furthermore, by combining it with a sentiment analysis engine, it achieves even more accurate prediction and optimization.

[1404] System configuration

[1405] This system consists of three main components: the server, the terminal, and the user. Each component will be explained in detail below.

[1406] Data collection and preprocessing

[1407] The server first collects customer purchasing information, inventory data, disposal data, and employee shift information. A relational database (e.g., MySQL or PostgreSQL) is used. After collecting the data, the server preprocesses the data and fills in missing and outlier values. Specifically, data cleaning is performed using the Python pandas library.

[1408] Demand forecasting and optimization

[1409] Based on the preprocessed data, the server uses a generative AI model (e.g., TensorFlow, PyTorch) to generate demand forecasts. Based on the generated demand forecasts, the server optimizes order quantities, optimizes pricing, and automatically generates employee shifts. For example, it uses a time series analysis model to forecast demand.

[1410] Sentiment analysis and feedback adjustment

[1411] The user checks the generated prediction and optimization results. Here, a sentiment analysis engine (e.g., OpenAI's GPT-4 model) analyzes the user's feedback and makes adjustments as necessary. Specifically, the user can check and correct the results using a dashboard provided in the web application.

[1412] Autorun and Notifications

[1413] The server automatically executes the necessary actions based on the information confirmed and corrected by the user. For example, if an order list is confirmed, the server automatically sends the order to the supplier using an API. It also distributes price information to the POS system.

[1414] Specific examples

[1415] A concrete example of how this works is shown below:

[1416] 1. Automated ordering system:

[1417] After analyzing data from the past year, it predicts that demand for a particular product will increase next week. Based on this prediction, an order list is generated and confirmed by the user. A sentiment analysis engine analyzes the user's reaction and adjusts the order quantity to the optimum. Finally, the server automatically sends the order list to the supplier.

[1418] 2. Dynamic Pricing System:

[1419] Based on past purchase history, it predicts high demand during certain times of the day and suggests pricing for those times. The user checks the pricing, and the sentiment analysis engine analyzes their reactions, readjusts the price, and automatically sends it to the POS system.

[1420] 3. Automatic shift creation:

[1421] If a high number of customers are predicted on certain weekends, a shift schedule is generated that increases the number of employees required during those times. Users can check the schedule, and the sentiment analysis engine analyzes reactions, adjusts it as necessary, and distributes it to employees.

[1422] 4. Service and product demand forecast:

[1423] Based on past data, certain products are predicted to be popular in the next month and are listed as promotional items. Users check the list, and a sentiment analysis engine analyzes reactions and makes adjustments, planning promotions, and replenishing inventory.

[1424] Example prompt sentence:

[1425] 1. "Based on the sales data from the past year, forecast next week's demand and calculate the optimal order quantity."

[1426] 2. "Based on past customer purchasing history, identify times of day when demand is high and suggest pricing for those times."

[1427] 3. "Generate optimal shift schedules based on employee shift history and store visit predictions."

[1428] 4. "Analyze sales data and create a recommended list of products to promote and replenish inventory for next month."

[1429] This completes the description of the embodiment of the invention. This system enables integrated data management and efficient business operations, and realizes highly accurate predictions and optimization that reflect user feedback.

[1430] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1431] Step 1: Data collection

[1432] The server retrieves customer purchasing information, inventory data, disposal data, and employee shift information from the database. SQL queries are used as input, and raw data extracted from the database is obtained as output. Specifically, the server executes periodically scheduled jobs to retrieve the required data.

[1433] Step 2: Data Preprocessing

[1434] The server preprocesses the acquired raw data. The input contains the raw data extracted in step 1, and the output is the preprocessed clean data. Specific operations include imputing missing values ​​(for example, median imputation) and correcting outliers using the pandas library. Outlier detection uses common statistical methods (for example, z-score).

[1435] Step 3: Demand forecast

[1436] The server uses a generative AI model (for example, a time series analysis model using TensorFlow) based on the preprocessed data to make a demand forecast. The input is the data preprocessed in step 2, and the output is the demand forecast result. Specifically, past purchase data is input into the model to predict future demand. In this process, the time series data is analyzed using a model such as LSTM.

[1437] Step 4: Optimize order quantity

[1438] The server optimizes the order quantity based on the demand forecast results. The input is the demand forecast results obtained in step 3, and the output is an optimized order list. Specifically, the demand forecast results are input into an optimization algorithm (e.g., linear programming) to calculate the order quantity.

[1439] Step 5: Optimize pricing

[1440] The server optimizes pricing based on the demand forecast results. The input is the demand forecast results obtained in step 3, and the output is an optimized price list. Specifically, it uses an AI model (e.g., regression analysis) to analyze the relationship between demand and price and sets prices.

[1441] Step 6: Automatic shift creation

[1442] The server automatically generates employee shifts based on the demand forecast results. The inputs are the demand forecast results obtained in step 3 and employee desired shift information, and the output is an optimal shift schedule. Specifically, the optimal shifts are generated using a shift generation algorithm (e.g., a Bayesian network).

[1443] Step 7: User confirmation and feedback

[1444] The user checks the generated order list, price list, and shift schedule. The input is the data provided by the server, and the output is the user's feedback. In concrete terms, the user checks the data on the web application, and the sentiment analysis engine analyzes the user's feedback. For example, the user may enter text feedback such as "This order quantity is too large."

[1445] Step 8: Analyze feedback and adjust

[1446] The server uses an emotion analysis engine to analyze user feedback and adjust prediction and optimization results as necessary. The input is user feedback, and the output is adjusted order lists, price lists, and shift schedules. Specifically, the emotion analysis engine analyzes the user's text and facial expressions and adjusts the data to reflect the results.

[1447] Step 9: Autorun and Notifications

[1448] The server automatically takes the necessary actions based on the adjusted information and notifies the relevant parties. The input is the data adjusted in step 8, and the output is the sent order list, updated price information, and distributed shift schedules. Specific operations include sending order information to suppliers using APIs, updating price information in the POS system, and distributing shift information to employees.

[1449] (Application example 2)

[1450] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1451] Conventional business operation systems have difficulty in efficiently optimizing demand forecasting, inventory management, pricing, and employee shift scheduling, and also have problems in that they do not respond to the burden and emotions felt by users. For these reasons, there is a need to efficiently operate business operations in physical stores and improve user satisfaction.

[1452] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1453] In this invention, the server includes means for collecting customer purchasing information, inventory data, disposal data, and employee shift information, means for preprocessing the collected data and complementing missing and outlier values, means for using an AI model based on the preprocessed data to perform demand forecasting, inventory order optimization, pricing optimization, and automatic employee shift generation, means for presenting necessary information to a user based on the prediction results and generated data and for confirmation and correction, means for automatically performing necessary actions based on the information confirmed and corrected by the user, means for collecting emotion data and adjusting information based on the user's emotion, and means for providing and managing information in real time via mobile terminals and wearable devices, thereby enabling smart, emotion-conscious business operations.

[1454] "Customer purchasing information" refers to data such as information about the products purchased by the customer, the date and time of purchase, and frequency of purchase.

[1455] "Inventory data" refers to data that indicates the current inventory status of products in stores and warehouses.

[1456] "Disposal data" refers to data relating to the disposal status of a product, the reason for disposal, and the date and time of disposal.

[1457] "Employee shift information" refers to information such as the work schedules and desired holidays of store employees.

[1458] "Preprocessing" is the process of complementing missing or outliers in collected data and preparing it in a form suitable for input into an AI model.

[1459] An "AI model" is an algorithm that uses machine learning and deep learning to predict demand and optimize data.

[1460] "Demand forecasting" is the process of predicting how much and how little a customer will purchase of a particular product or service.

[1461] "Optimizing inventory order quantities" means determining the appropriate order quantity, neither too much nor too little, based on demand forecasts.

[1462] "Pricing optimization" means setting the optimal selling price for a product, taking into account supply and demand conditions.

[1463] "Automatic generation of employee shifts" means automatically creating optimal shift schedules by taking into account store visitor forecasts and employees' desired holidays.

[1464] "Emotion data" is data based on the recognition of the user's reactions and emotions.

[1465] "Mobile terminals and wearable devices" refers to portable devices such as smartphones, smart glasses, and head-mounted displays.

[1466] "Means for providing and managing information in real time" refers to a method for obtaining data instantly, providing information to users, and efficiently managing business operations.

[1467] This invention provides a system for optimizing the business operations of a physical store based on customer purchase information, inventory data, disposal data, and employee shift information. The operation of the system is described in detail below.

[1468] System configuration

[1469] This system mainly consists of four main components: a server, a mobile terminal, a wearable device, and a user. The roles of each are as follows:

[1470] Server Roles

[1471] The server comprehensively manages customer purchasing information, inventory data, disposal data, and employee shift information. Specifically, it collects and preprocesses this data and applies AI models to perform the following tasks:

[1472] Data collection and preprocessing

[1473] The server automatically collects various data and performs preprocessing to fill in missing and outlier values. Specific software used is the Python pandas library and scikit-learn.

[1474] Demand Forecasting and Optimization

[1475] Based on the preprocessed data, an AI model (for example, a deep learning model using TensorFlow or Keras) is used to predict demand for the following week and optimize order quantities and pricing.

[1476] Emotion data capture

[1477] It collects user emotional data and adjusts prediction results based on this data. The emotion engine uses an NLP model to analyze emotions from user text and voice input.

[1478] Autorun

[1479] Based on the information confirmed and corrected by the user, necessary actions (such as sending an order list or updating a shift schedule) are automatically performed.

[1480] Mobile and wearable devices

[1481] Mobile devices (such as smartphones) and wearable devices (such as smart glasses and head-mounted displays) serve as tools to provide users with information in real time and assist in management.

[1482] Real-time information provision

[1483] Inventory information, sales status, employee shift information, and more are instantly presented to users via mobile devices or wearable devices.

[1484] User Interface

[1485] These devices allow for voice and touch input as user interfaces, allowing users to easily check and modify information.

[1486] User operations

[1487] Users can check the information provided in real time and adjust order lists, pricing, and shift schedules as needed. These adjustments are fed back to the emotion engine and reflected in optimization.

[1488] Component Examples

[1489] A concrete example is an inventory management assistant for a brick-and-mortar store, where store staff can use smart glasses to check inventory information in real time, view items that need to be ordered, and forecast demand for specific times of day. The assistant also uses an emotion engine to collect staff emotions and automatically adjust forecasts accordingly.

[1490] Prompt Sentence Examples

[1491] To manage your store's inventory in real time, please provide us with a dataset that includes the following items:

[1492] Product ID

[1493] Product name

[1494] Current Stock

[1495] Inventory quantity one week ago

[1496] Disposal Data

[1497] Past sales data

[1498] In this way, collecting emotional data and optimizing business operations will enable efficient business operations in physical stores.

[1499] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1500] Step 1:

[1501] The server collects customer purchasing information, inventory data, waste data, and employee shift information through APIs. To obtain this data, the server uses HTTP requests to retrieve the necessary information from databases and external services. The input data is in JSON or CSV format, and the output is the collected raw data.

[1502] Step 2:

[1503] The server preprocesses the collected raw data. Specifically, it converts it into a data frame using the pandas library and imputes missing or outlier values. It also normalizes and standardizes the data as needed. This results in data in a format suitable for input to the AI ​​model. The input is the collected raw data, and the output is the preprocessed data.

[1504] Step 3:

[1505] The server applies AI models to the preprocessed data to forecast demand and optimize inventory order quantities. Specifically, it loads trained models using TensorFlow and Keras and executes predictions. The input is the preprocessed data, and the output is the prediction results.

[1506] Step 4:

[1507] The server collects the user's emotional data. The emotional data is obtained by analyzing the user's text and voice input with an NLP model. Specifically, it uses Hugging Face's Transformers for emotion analysis. The input is the user's text and voice data, and the output is the analyzed emotional data.

[1508] Step 5:

[1509] The server adjusts the prediction results based on the analyzed emotional data. By incorporating emotional data into the prediction results, more realistic and appropriate results can be obtained. For example, if there is a lot of negative emotional feedback, the demand forecast volume can be reduced. The inputs are the prediction results and emotional data, and the output is the adjusted prediction results.

[1510] Step 6:

[1511] The mobile terminal or wearable device provides the adjusted prediction results to the user in real time. The user can check the information through the device and make corrections as needed. Corrections are made via voice or touch input, and the data is sent back to the server. The input is the adjusted prediction results, and the output is the corrected prediction results or user feedback.

[1512] Step 7:

[1513] The server automatically executes the final action based on the information modified by the user, such as sending a final inventory order list to suppliers or notifying employees of shift schedules. The input is the modified prediction results and feedback, and the output is the specific action to be taken.

[1514] Through the above steps, the server, mobile terminal, wearable device, and user can work together to optimize the business operations of a physical store and improve user satisfaction.

[1515] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1516] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1517] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1518] [Fourth embodiment]

[1519] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1520] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1521] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1522] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1523] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1524] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1525] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1526] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1527] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1528] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1529] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1530] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1531] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1532] This invention relates to a system for optimizing specific business operations, and aims to improve business efficiency by applying AI models to mainly utilize customer purchasing information, inventory data, waste data, and employee shift information. The system consists of three main components: a server, a terminal, and a user.

[1533] Automated ordering system

[1534] Server: In the ordering system, the server first collects inventory status, past order data, and waste data. Next, the server preprocesses this data and fills in missing and outlier values. Based on the preprocessed data, the server uses an AI model to forecast demand and calculate the next order quantity.

[1535] For example, if an analysis of data from the past year predicts that demand for a particular product will increase next week, the server automatically generates an appropriate order quantity based on that prediction. After the user checks this order list and makes any necessary corrections, the server automatically sends the final list to the supplier.

[1536] Dynamic Pricing System

[1537] Server: In this system, the server collects customer purchase history and store visits by time of day, and pre-processes this data. Based on the pre-processed data, the server uses an AI model to analyze the relationship between demand and price. Next, it sets the optimal price based on demand and automatically distributes it to the POS system.

[1538] For example, if past purchase history shows that demand is highest between 3:00 PM and 6:00 PM on weekdays, you can set the price to be 10% higher during that time period, which will reduce unsold items and maximize profits.

[1539] Automatic shift creation

[1540] Server: In the shift creation system, the server collects employee shift history, desired vacation time, and store visit predictions. It then preprocesses this data and uses an AI model to predict optimal shifts based on the number of employees and peak hours. The predicted shift schedule is automatically generated and can be reviewed and modified by the user.

[1541] For example, if a high number of customers is predicted on a particular weekend, the system generates a shift schedule that automatically allocates the necessary number of employees during that time period. The user can check the generated schedule, correct any problems, and automatically distribute it to employees.

[1542] Service and product demand forecasting

[1543] Server: In a demand forecasting system, the server collects and preprocesses customer purchase history and sales data. Based on the preprocessed data, the server uses an AI model to predict future demand. Based on the prediction results, it generates a list of recommended promotional items and inventory replenishment and presents it to the user.

[1544] For example, if past data predicts that a particular product will be popular in the next month, the product will be listed as a promotion target. Users can check the list and make appropriate promotion plans and inventory replenishment.

[1545] In this way, the invention will reduce food waste, alleviate labor shortages, and achieve sustainable service provision through appropriate pricing.

[1546] The processing flow will be explained below.

[1547] Automated ordering system processing

[1548] Step 1:

[1549] Server: Collects inventory status, past order data, and disposal data from the database.

[1550] Every day at 2:00 AM, the server collects inventory status, order data, and disposal data from the database for the past year.

[1551] Step 2:

[1552] Server: Preprocesses the collected data and imputes missing or outlier values.

[1553] After collecting the data, the server performs preprocessing, imputing missing values ​​with the average value and removing outliers.

[1554] Step 3:

[1555] Server: Preprocessed data is input into the AI ​​model to generate demand forecasts.

[1556] The server passes the pre-processed data to an AI model to generate demand forecasts for the next week.

[1557] Step 4:

[1558] Server: Calculates the optimal order quantity based on the demand forecast results.

[1559] The server calculates the optimal order quantity for each product based on the demand forecast results output by the AI ​​model.

[1560] Step 5:

[1561] Server: Automatically generate an order list based on the optimal order quantity.

[1562] The server creates an order list based on the calculated order quantity.

[1563] Step 6:

[1564] User: Check the generated order list and modify it as necessary.

[1565] The user checks the order list on the system and corrects the quantity if necessary.

[1566] Step 7:

[1567] Server: Automatically sends the confirmed and corrected order list to the supplier.

[1568] The server receives the order list that the user has confirmed and modified, and automatically sends it to the specified supplier.

[1569] Dynamic pricing system processing

[1570] Step 1:

[1571] Server: Collects customer purchase history and store visit data by time of day.

[1572] Every night, the server collects customer purchase history and time-of-day visit data from the database.

[1573] Step 2:

[1574] Server: Preprocesses the collected data and removes outliers.

[1575] The server performs preprocessing after data collection to normalize or remove outliers.

[1576] Step 3:

[1577] Server: Inputs the preprocessed data into the AI ​​model and analyzes the relationship between demand and price.

[1578] The server passes the pre-processed data to an AI model that analyzes the relationship between demand and price.

[1579] Step 4:

[1580] Server: Sets the price for the next day based on the analysis results.

[1581] The server uses the analysis results output by the AI ​​model to set the optimal price for the next day.

[1582] Step 5:

[1583] Server: Automatically distributes the set price information to the POS system.

[1584] The server sends the set price information to the POS system, which updates it automatically.

[1585] Process automatic shift creation

[1586] Step 1:

[1587] Server: Collects employee shift history, desired vacation time, and store visit forecast data.

[1588] At the end of each month, the server collects employee shift history, desired vacation data, and store visit forecast data from the database.

[1589] Step 2:

[1590] Server: Preprocesses the collected data and imputes missing or outlier values.

[1591] After data collection, the server performs preprocessing, imputing missing values ​​and removing outliers.

[1592] Step 3:

[1593] Server: Based on the pre-processed data, the AI ​​model predicts the optimal shift.

[1594] The server inputs the pre-processed data into an AI model to predict the optimal shift schedule.

[1595] Step 4:

[1596] Server: Automatically generates predicted shift schedules.

[1597] The server automatically generates shift schedules based on the prediction results output by the AI ​​model.

[1598] Step 5:

[1599] User: Review the generated shift schedule and modify it if necessary.

[1600] Users can check the shift schedule on the system and modify it if necessary.

[1601] Step 6:

[1602] Server: Automatically distributes confirmed shift schedules to employees.

[1603] The server receives the shift schedule that the user has confirmed and modified and automatically distributes it to all employees.

[1604] Processing demand forecasts for services and products

[1605] Step 1:

[1606] Server: Collects customer purchase history and sales data.

[1607] The server collects customer purchase history and sales data from the database every day at 11:00 p.m.

[1608] Step 2:

[1609] Server: Preprocesses the collected data and imputes missing or outlier values.

[1610] After data collection, the server performs preprocessing, imputing missing values ​​and removing outliers.

[1611] Step 3:

[1612] Server: Inputs the pre-processed data into the AI ​​model to predict future demand.

[1613] The server passes the pre-processed data to an AI model to predict future demand.

[1614] Step 4:

[1615] Server: Based on the prediction results, a list of recommended promotion targets and inventory replenishment is generated.

[1616] The server automatically generates a list of recommended promotional targets and inventory replenishment based on the prediction results output by the AI ​​model.

[1617] Step 5:

[1618] User: Review the generated recommendation list and incorporate it into promotion planning and inventory replenishment.

[1619] The user checks the recommended list and implements promotion plans and inventory replenishment as necessary.

[1620] Example 1

[1621] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1622] In conventional business systems, customer purchasing information, inventory data, disposal data, and employee shift information are often managed separately, making it difficult to integrate and efficiently utilize this data. Furthermore, business processes such as demand forecasting, pricing, and automatic shift generation are performed manually, requiring a great deal of time and effort, making it difficult to improve the efficiency and accuracy of business operations. Aiming to solve these problems, the present invention provides a system for optimizing specific business operations.

[1623] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1624] In this invention, the server includes means for collecting customer purchasing information, inventory data, disposal data, and employee shift information, means for preprocessing the collected data and complementing missing values ​​and outliers, means for using a generative AI model based on the preprocessed data to perform demand forecasting, optimizing inventory order quantities, optimizing pricing, and automatically generating employee shifts, means for presenting necessary information to a user based on the prediction results and generated data and for confirmation and correction, and means for automatically executing necessary actions based on the information confirmed and corrected by the user. This enables efficient management of customer purchasing information, inventory data, disposal data, and employee shift information, automating business processes such as demand forecasting, inventory ordering, pricing, and shift creation, and improving business efficiency and accuracy.

[1625] Key Word Definitions

[1626] "Customer purchasing information" refers to data that records the type, quantity, purchase date, etc. of products that a customer has purchased in the past.

[1627] "Inventory data" refers to data for managing the quantity of products currently in stock, storage locations, and inventory history.

[1628] "Disposal data" refers to data that records the quantity of products that have been discarded due to reasons such as expiration or poor quality, the reason for disposal, and the date and time of disposal.

[1629] "Employee shift information" refers to data necessary for labor management, such as employees' working hours, desired holidays, and shift history.

[1630] "Missing values" refers to a state in which data that should be present in a dataset is missing.

[1631] An "outlier" is a value that is significantly different from other data in a data set and is typically treated as an error or outlier.

[1632] A "generative AI model" is a model that uses artificial intelligence technology to learn specific patterns from data and automatically perform tasks such as prediction and classification.

[1633] "Demand forecasting" is the process of predicting future demand based on past data, and refers to predicting sales quantities, usage amounts, etc.

[1634] "Inventory order optimization" is the process of determining optimal inventory levels based on demand forecasts to avoid shortages and excess inventory.

[1635] "Pricing optimization" is the process of setting optimal selling prices taking into account demand and market conditions, with the goal of maximizing profits.

[1636] "Automatic employee shift generation" is the process of automatically creating employee work schedules, taking into account business demands and employee preferences.

[1637] "Means for automatically taking necessary actions based on information confirmed and corrected by the user" refers to a process for automatically taking business actions such as placing orders, changing prices, and notifying shifts based on data confirmed and corrected by the user.

[1638] MODE FOR CARRYING OUT THE INVENTION

[1639] This invention relates to a system for optimizing specific business operations, and aims to improve business efficiency by applying generative AI models primarily using customer purchasing information, inventory data, waste data, and employee shift information. This system consists of three main components: a server, a terminal, and a user.

[1640] Hardware and software used

[1641] The server is built on AWS (a common cloud computing service) and uses MySQL as the database. Python and libraries such as Pandas and Scikit-learn are used for data preprocessing and analysis. AI frameworks such as TensorFlow, Keras, XGBoost, and Prophet are mainly used to implement the generative AI model. A general sales management system is assumed as the POS system.

[1642] System example and operation

[1643] Automated ordering system

[1644] Server: The server periodically collects inventory status, past order data, and waste data from terminals and point-of-sale (POS) systems, and preprocesses this data using Python and Pandas. It imputes missing and outlier values ​​and inputs the data into a generative AI model (e.g., TensorFlow) to perform demand forecasting. It automatically generates an order list based on the forecast results and presents it to the user. The user can review and modify the order list, and the final list is automatically sent from the server to the supplier.

[1645] Example: Based on data from the past year, the order quantity for the next week, when demand is predicted to increase, is automatically calculated and an order list is created.

[1646] Dynamic Pricing System

[1647] Server: The server collects and preprocesses customer purchase history and store visit status by time of day in real time. This data is input into a generative AI model (e.g., XGBoost) to analyze the relationship between demand and price. The optimal price is calculated from the analysis results and automatically distributed to the sales management system.

[1648] Example: If analysis shows that demand is high between 3pm and 6pm on weekdays, the price during that time period will be increased by 10%.

[1649] Automatic shift creation

[1650] Server: The server collects and preprocesses employee shift history, desired vacation time, and store visit prediction information. This data is input into a generative AI model (e.g., Keras) to predict and automatically generate employee shift schedules. Users can check and modify the generated shift schedules. The final shift schedules are automatically distributed from the server to employees.

[1651] Example: If store visits are predicted to be high on a particular weekend, create a shift that places more employees at that time.

[1652] Service and product demand forecasting

[1653] Server: The server periodically collects customer purchase history and sales data, and after preprocessing, inputs it into a generative AI model (e.g., Prophet) to predict future demand. Based on the prediction results, it generates a list of recommended promotional items and inventory replenishment and presents it to the user. The user checks this list and takes the necessary actions.

[1654] Example: If historical data predicts that a particular product will be popular next month, target that product for promotion.

[1655] Prompt Sentence Examples

[1656] "Please forecast demand for the next month based on data from the past year, and propose periods when demand will be high and product promotions. Please also let us know specific promotion ideas."

[1657] In this way, the present invention is a system that uses a generative AI model to integrate and analyze a variety of business data and automatically implement optimized business processes, thereby improving business efficiency and supporting highly accurate decision-making.

[1658] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1659] Processing steps of specific business operation optimization system

[1660] Automated ordering system processing flow

[1661] Step 1: Data collection

[1662] Server: The server periodically collects inventory status, past order data, and waste data from terminals and POS systems. Raw data obtained from each terminal is sent to the server as input. The output is consolidated data stored in a database within the server. As a concrete example, an automatic data collection script is executed every night.

[1663] Step 2: Data Preprocessing

[1664] Server: The collected data is preprocessed using Python and Pandas. The collected raw data is used as input. Missing values ​​and outliers are imputed, and the output is formatted as time series data. For example, missing values ​​are imputed with the past average value, and outliers with the median value.

[1665] Step 3: Demand forecasting using AI models

[1666] Server: The preprocessed data is input into a generative AI model (e.g., TensorFlow) to generate a demand forecast. The forecast model takes into account trends and seasonal fluctuations based on past data. The preprocessed data is used as input, and a forecast of future demand is output. For example, if demand for next week is predicted to be 200 units, the server outputs that number.

[1667] Step 4: Generate an order list

[1668] Server: Automatically generates an order list based on the forecast results. Based on the forecasted demand, the optimal order quantity is calculated taking into account the safety stock amount, and compiled in list format. The demand forecast results are used as input, and the output is an order list. As a specific example, if the forecasted demand is 200 units, an order list for 220 units is generated taking into account the safety stock amount.

[1669] Step 5: Check and modify the order list

[1670] User: The user checks the generated order list in a dedicated interface and modifies it if necessary. The automatically generated order list is used as input, and a modified final list is generated as output. For example, if the user feels that there are too many of a particular product in this list, the order quantity can be adjusted by the user.

[1671] Step 6: Send the order list to your suppliers

[1672] Server: Automatically sends the final purchase order list, which has been reviewed and modified by the user, to the supplier. It uses the modified purchase order list as input and generates purchase instructions as output, which are sent to the supplier, for example, via email or electronic data interchange (EDI).

[1673] Dynamic pricing system processing flow

[1674] Step 1: Data collection

[1675] Server: The server collects customer purchase history and store visit status by time of day in real time. The input is raw data obtained from the POS system, and the output is purchase data stored in a database within the server. For example, the latest data is obtained from the POS system every hour.

[1676] Step 2: Data Preprocessing

[1677] Server: Cleanses the collected data and removes outliers using Python and Scikit-learn. The input is raw data collected from the POS system, and the output is a clean dataset. For example, the format of dates and times is standardized.

[1678] Step 3: Analyze the relationship between demand and price

[1679] Server: The preprocessed data is input into an AI model (e.g., XGBoost) to analyze the relationship between demand and price. The input is the preprocessed data, and the output is the analysis result. For example, demand forecasting is performed using the relationship between past price fluctuations and demand.

[1680] Step 4: Set the optimal price

[1681] Server: Sets the optimal price based on the analysis results. The analysis results are used as input and new pricing is generated as output. For example, the price is set to increase by 10% during times of high demand.

[1682] Step 5: Distributing price information to POS systems

[1683] Server: Calculated price information is sent to the POS system in real time, automatically updating the store's prices. The input is the new price setting, and the output is the price information reflected in the POS system. For example, the new price information is immediately applied to the POS system and displayed to the customer.

[1684] Process flow for automatic shift creation

[1685] Step 1: Data collection

[1686] Server: The server collects employee shift history, desired vacation time, and store visit prediction information. The employee database and store visit prediction data are used as input, and the output is the integrated data stored in the server. For example, the latest data is obtained every Sunday.

[1687] Step 2: Data Preprocessing

[1688] Server: Cleanses the collected data and processes missing and outliers using Python and Pandas. The input is raw data, and the output is a clean dataset. For example, missing values ​​are imputed with the mean and outliers are removed.

[1689] Step 3: Shift prediction using AI models

[1690] Server: Input the preprocessed data into an AI model (e.g., Keras) to predict employee shifts. The input is the preprocessed data, and the output is the predicted shift schedule. For example, it allocates the required number of employees according to peak hours.

[1691] Step 4: Generate a shift schedule

[1692] Server: Automatically generates shift schedules based on the prediction results. The input is the prediction results, and the output is the generated shift schedule. As a specific example, shifts are adjusted according to the predicted number of customers.

[1693] Step 5: Check and modify the shift schedule

[1694] User: The user checks the generated shift schedule and modifies it if necessary. The input is the automatically generated shift schedule, and the output is the final modified schedule. For example, if a user requests sudden leave, the shift can be covered by another employee.

[1695] Step 6: Distribute shift schedules to employees

[1696] Server: Automatically distributes the final shift schedule to employees. The input is the revised shift schedule, and the output is the shift information sent to employees. For example, it can be sent via email or a dedicated app.

[1697] Service / product demand forecasting process flow

[1698] Step 1: Data collection

[1699] Server: The server periodically collects customer purchase history and sales data. The input is raw data obtained from the POS system, and the output is sales data stored in a database on the server. For example, data is collected at the end of each month.

[1700] Step 2: Data Preprocessing

[1701] Server: Cleanses the collected data and processes missing values ​​and outliers using Python and Pandas. The input is raw data, and the output is a clean dataset. For example, missing values ​​are imputed with the average value of the previous month, and outliers are removed.

[1702] Step 3: Demand forecasting using AI models

[1703] Server: Inputs the preprocessed data into a generative AI model (e.g., Prophet) to predict future demand. The input is the preprocessed data, and the output is the prediction result. For example, a specific product is expected to be in high demand next month.

[1704] Step 4: Generate a recommendation list

[1705] Server: Based on the prediction results, a list of recommendations for promotions and inventory replenishment is generated. The input is the prediction results, and the output is the generated recommendation list. As a concrete example, it lists products with increasing demand.

[1706] Step 5: Review and modify the recommendation list

[1707] User: The user reviews the generated recommendation list and modifies it if necessary. The input is the automatically generated recommendation list, and the output is the modified list. For example, the user determines that a critical item needs to be restocked.

[1708] Step 6: Planning promotions and replenishing inventory

[1709] User: After reviewing the recommendation list, create promotion plans or replenish inventory based on the list. The input is the revised recommendation list, and the output is the implemented promotions or replenished inventory. For example, target specific products for a special campaign.

[1710] (Application example 1)

[1711] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1712] In traditional brick-and-mortar stores, inventory management, pricing, demand forecasting, and employee shift planning are often performed manually, resulting in problems such as inefficiency and human error in each process. Furthermore, there was a lack of means to grasp trends in customer purchasing behavior in real time and respond quickly, making appropriate inventory management and pricing difficult. Furthermore, it was difficult to implement automatic ordering based on forecasts or dynamic pricing to prevent unsold items, making it difficult to develop effective sales strategies. It is essential to solve these problems and improve the overall efficiency of store operations.

[1713] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1714] In this invention, the server includes: means for collecting customer purchasing information, inventory data, disposal data, and employee shift information; means for preprocessing the collected data and complementing missing and outlier values; means for using an AI model based on the preprocessed data to forecast demand, optimize inventory order quantities, optimize pricing, and automatically generate employee shifts; means for presenting necessary information to a user based on the forecast results and generated data and allowing the user to confirm and correct it; means for automatically executing necessary actions based on the information confirmed and corrected by the user; means for providing inventory management functions in real time and issuing alerts to prevent stockouts and excess inventory; means for dynamically setting product prices based on purchase history and the number of customers by time period to prevent unsold items and maximize profits; means for automatically creating optimal shift schedules based on employee shift history and desired vacation days; and means for predicting future demand based on customer purchase history and sales data and generating recommended lists of promoted products and inventory replenishment. This allows for centralized management of various processes related to store operations, enabling efficient and accurate inventory management, pricing, demand forecasting, and shift creation.

[1715] "Customer purchasing information" is data including the type and quantity of products purchased by customers who visit a store, the date and time of purchase, and the attributes of the purchaser.

[1716] "Inventory data" refers to information relating to the current quantity of products held in the store, storage locations, replenishment status, and the like.

[1717] "Disposal data" is data indicating the type, quantity, and date and time of disposal of products that have been discarded due to the expiration of their sales period or deterioration of quality.

[1718] "Employee shift information" is data including an employee's working date and time, working hours, vacation days, desired work shifts, etc.

[1719] "Preprocessing" is the process of converting collected data into a format suitable for analysis and model training, and filling in missing or outlier values.

[1720] An "AI model" is a mathematical model that uses artificial intelligence technologies such as machine learning and deep learning to make useful predictions and classifications from data.

[1721] "Demand forecasting" is the process of predicting future demand for a product based on past data.

[1722] "Inventory order quantity optimization" is the process of preventing excess inventory and out-of-stock by predicting product demand and calculating the appropriate order quantity based on that.

[1723] "Pricing optimization" is the process of adjusting the price of a product to an optimal value based on demand data, etc.

[1724] "Automatic generation of employee shifts" is the process of automatically creating optimal shift schedules based on employees' shift history, desired working hours, and store business demands.

[1725] "Prediction results and generated data" refers to information such as demand forecast results, optimal order quantities, appropriate prices, and shift schedules generated by AI models.

[1726] The "inventory management function" monitors inventory status in real time and reduces the risk of stockouts and excess inventory.

[1727] "Dynamic setting" means that values ​​and parameters can be flexibly changed according to real-time data and the current situation.

[1728] "Promotional Products" means selected products that are specially promoted or discounted for promotional purposes.

[1729] A "recommended list" is a list generated by the system to encourage specific actions or responses.

[1730] This invention is built as a system for optimizing specific business operations. The system consists of three main components: a server, a terminal, and a user.

[1731] Inventory management function

[1732] The server first collects customer purchasing information, inventory data, waste data, and employee shift information. This data is obtained from databases and POS systems. The server then preprocesses the data and fills in missing and outlier values. Specifically, it cleanses the data using data analysis libraries such as Pandas. Using the preprocessed data, it trains a demand forecasting model using AI libraries such as TensorFlow and Keras. This model predicts future inventory demand and calculates appropriate order quantities.

[1733] Dynamic Pricing Feature

[1734] The server also analyzes purchase history and store visits by time of day to determine optimal pricing based on demand. This data is also preprocessed and fed into the AI ​​model, which analyzes the relationship between demand and price and calculates the most profitable price. This price information is automatically sent to the POS system, where prices are adjusted in real time.

[1735] Automatic shift creation

[1736] The server automatically generates optimal shift schedules based on employees' shift history, desired holidays, and predicted store visits. Employee information and store visit prediction data are fed into an AI model to calculate the optimal shift schedule. Users can check and modify the generated schedule, and the final shift schedule is distributed to employees.

[1737] Demand forecasting and promotions

[1738] The server also collects customer purchase history and sales data to predict future demand. Based on this prediction, it targets specific products for promotion and generates a list of recommended inventory replenishment. Specifically, if the AI ​​model predicts that a particular product will be popular in the next month, it will list that product as a promotion target.

[1739] Hardware and software used

[1740] The hardware used is a server for database hosting and AI model training, and a smartphone as a client terminal. The software used is Pandas (data collection and preprocessing) and TensorFlow / Keras (demand forecasting AI model construction and training).

[1741] Examples of specific examples and prompts

[1742] For example, if a particular product is predicted to be popular in the coming month, the product may be listed for promotion and the store may be notified.

[1743] Example prompt sentence:

[1744] "Using past sales data, the AI ​​model predicts which products will be in high demand for the next month, and creates a promotion plan for those products."

[1745] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1746] Step 1:

[1747] The server collects customer purchasing information, inventory data, disposal data, and employee shift information.

[1748] These data are obtained from databases and POS systems. The collected data is first converted into a data frame using the Pandas library, and missing and outlier values ​​are imputed (data cleansing).

[1749] Input: Customer purchasing information, inventory data, disposal data, employee shift information

[1750] Output: Preprocessed data frame

[1751] Step 2:

[1752] The server trains the AI ​​model based on the preprocessed data.

[1753] Build and train a demand forecasting model using TensorFlow and Keras libraries. Historical purchase and inventory data are used as input data for the model to generate demand forecasts.

[1754] Input: Preprocessed data frame

[1755] Output: A trained demand forecasting model

[1756] Step 3:

[1757] The server uses the trained AI model to predict future inventory demand and calculate appropriate order quantities.

[1758] Past purchasing data is input into the forecasting model to predict demand for the next week or month. Based on the forecast results, the optimal inventory order quantity is calculated.

[1759] Inputs: Trained demand forecasting model, historical purchase data

[1760] Output: Future inventory demand forecast, order quantity

[1761] Step 4:

[1762] The server automatically generates an order list based on the demand forecast results and the order quantity, and presents it to the user.

[1763] The generated order list is displayed on the terminal for the user to review and modify, and after the user makes the appropriate modifications, the information is automatically sent to the supplier.

[1764] Input: Future inventory demand forecast, order quantity

[1765] Output: Order list, order data to suppliers

[1766] Step 5:

[1767] The server analyzes purchase history and store visits by time of day, and sets optimal prices based on demand.

[1768] The AI ​​model analyzes the relationship between demand and price to calculate the optimal price, which is then automatically sent to the POS system.

[1769] Input: Purchase history, store visits by time period

[1770] Output: Best price, price information to POS system

[1771] Step 6:

[1772] The server automatically generates an optimal shift schedule based on employee shift history, desired holidays, and predicted store visits, and presents it to the user.

[1773] The AI ​​model calculates the optimal shift schedule, which the user can then review and modify. The modified schedule is then distributed to employees.

[1774] Input: Shift history, desired leave, store visit forecast

[1775] Output: Shift schedule, information distributed to employees

[1776] Step 7:

[1777] The server predicts future demand based on customer purchase history and sales data, generates a recommended list of promotional items and inventory replenishment, and presents it to the user.

[1778] Based on the generated recommendation list, the user plans promotions and stock replenishment.

[1779] Input: Purchase history, sales data

[1780] Output: Demand forecast, recommendation list, promotional products

[1781] The above are the specific processing steps for carrying out the present invention.

[1782] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1783] This invention relates to a system for optimizing specific business operations, primarily using customer purchasing information, inventory data, waste data, and employee shift information to apply AI models to improve business efficiency. Furthermore, by combining it with an emotion engine that recognizes user emotions, even more accurate predictions and optimization can be achieved. This system consists of three main components: a server, a terminal, and a user.

[1784] Automated ordering system

[1785] Server: In the ordering system, the server first collects inventory status, past order data, and waste data. Next, the server preprocesses this data and fills in missing and outlier values. Based on the preprocessed data, the server uses an AI model to forecast demand and calculate the next order quantity. It also uses an emotion engine to incorporate user feedback and adjust the optimal order quantity.

[1786] For example, if an analysis of data from the past year predicts that demand for a particular product will increase next week, the system automatically generates an appropriate order quantity based on that prediction. After the user reviews this order list and the emotion engine analyzes the user's reactions and makes any necessary adjustments, the server automatically sends the final list to the supplier.

[1787] Dynamic Pricing System

[1788] Server: In this system, the server collects customer purchase history and store visits by time of day, and pre-processes this data. Based on the pre-processed data, the server uses an AI model to analyze the relationship between demand and price. Next, the server sets the optimal price based on demand, and an emotion engine incorporates user feedback to adjust the price setting, which is then automatically distributed to the POS system.

[1789] For example, if past purchasing history shows that demand is high between 3:00 PM and 6:00 PM on weekdays, the price can be increased by 10% during that time period, reducing unsold items and maximizing profits. The emotion engine analyzes user reactions and readjusts prices based on feedback.

[1790] Automatic shift creation

[1791] Server: In the shift creation system, the server collects employee shift history, desired vacation time, and store visit predictions. It then preprocesses this data and uses an AI model to predict optimal shifts based on the number of employees and peak hours. The predicted shift schedule is automatically generated and confirmed by the user. An emotion engine also analyzes the user's reactions and adjusts the shift schedule as necessary.

[1792] As a specific example, if a high number of customers is predicted on a particular weekend, the system will automatically generate a shift schedule that allocates the necessary number of employees during that time period. The user can check the generated schedule, and the emotion engine will analyze the user's reactions, correct any problems, and automatically distribute the schedule to employees.

[1793] Service and product demand forecasting

[1794] Server: In a demand forecasting system, the server collects and preprocesses customer purchase history and sales data. Based on the preprocessed data, the server uses an AI model to predict future demand. Based on the prediction results, it generates a list of recommended promotional items and inventory replenishment and presents it to the user. An emotion engine analyzes user reactions and adjusts the recommended list.

[1795] For example, if past data predicts that a particular product will be popular in the next month, the product will be listed as a promotion target. Users can check the list, and the emotion engine will analyze their reactions and make appropriate adjustments, such as planning promotions and replenishing inventory.

[1796] In this way, the invention will realize sustainable service provision through food waste reduction, labor shortage mitigation, and appropriate pricing. In addition, by integrating an emotion engine, more accurate prediction and optimization will be possible, leading to improved user satisfaction.

[1797] The processing flow will be explained below.

[1798] Automated ordering system processing

[1799] Step 1:

[1800] Server: Collects inventory status, past order data, and disposal data from the database.

[1801] Every day at 2:00 AM, the server collects inventory status, order data, and disposal data from the database for the past year.

[1802] Step 2:

[1803] Server: Preprocesses the collected data and imputes missing or outlier values.

[1804] After collecting the data, the server performs preprocessing, imputing missing values ​​with the average value and removing outliers.

[1805] Step 3:

[1806] Server: Preprocessed data is input into the AI ​​model to generate demand forecasts.

[1807] The server passes the pre-processed data to an AI model to generate demand forecasts for the next week.

[1808] Step 4:

[1809] Server: Calculates the optimal order quantity based on the demand forecast results.

[1810] The server calculates the optimal order quantity for each product based on the demand forecast results output by the AI ​​model.

[1811] Step 5:

[1812] Server: Automatically generate an order list based on the optimal order quantity.

[1813] The server creates an order list based on the calculated order quantity.

[1814] Step 6:

[1815] User: Check the generated order list and modify it as necessary.

[1816] The user checks the order list on the system and corrects the quantity if necessary.

[1817] Step 7:

[1818] Emotion Engine: Analyzes user reactions and suggests adjustments to the order list.

[1819] The emotion engine analyzes user reactions in real time and suggests adjustments to order quantities.

[1820] Step 8:

[1821] Server: Automatically sends the confirmed and corrected order list to the supplier.

[1822] The server receives the order list that the user has confirmed and modified, and automatically sends it to the specified supplier.

[1823] Dynamic pricing system processing

[1824] Step 1:

[1825] Server: Collects customer purchase history and store visit data by time of day.

[1826] Every night, the server collects customer purchase history and time-of-day visit data from the database.

[1827] Step 2:

[1828] Server: Preprocesses the collected data and removes outliers.

[1829] The server performs preprocessing after data collection to normalize or remove outliers.

[1830] Step 3:

[1831] Server: Inputs the preprocessed data into the AI ​​model and analyzes the relationship between demand and price.

[1832] The server passes the pre-processed data to an AI model that analyzes the relationship between demand and price.

[1833] Step 4:

[1834] Server: Sets the price for the next day based on the analysis results.

[1835] The server uses the analysis results output by the AI ​​model to set the optimal price for the next day.

[1836] Step 5:

[1837] Emotion Engine: Analyzes user reactions and suggests pricing adjustments.

[1838] The sentiment engine analyzes user reactions in real time and suggests pricing adjustments.

[1839] Step 6:

[1840] Server: Automatically distributes the set price information to the POS system.

[1841] The server sends the set price information to the POS system, which updates it automatically.

[1842] Process automatic shift creation

[1843] Step 1:

[1844] Server: Collects employee shift history, desired vacation time, and store visit forecast data.

[1845] At the end of each month, the server collects employee shift history, desired vacation data, and store visit forecast data from the database.

[1846] Step 2:

[1847] Server: Preprocesses the collected data and imputes missing or outlier values.

[1848] After data collection, the server performs preprocessing, imputing missing values ​​and removing outliers.

[1849] Step 3:

[1850] Server: Based on the pre-processed data, the AI ​​model predicts the optimal shift.

[1851] The server inputs the pre-processed data into an AI model to predict the optimal shift schedule.

[1852] Step 4:

[1853] Server: Automatically generates predicted shift schedules.

[1854] The server automatically generates shift schedules based on the prediction results output by the AI ​​model.

[1855] Step 5:

[1856] User: Review the generated shift schedule and modify it if necessary.

[1857] Users can check the shift schedule on the system and modify it if necessary.

[1858] Step 6:

[1859] Emotion engine: Analyzes user reactions and suggests shift schedule adjustments.

[1860] The emotion engine analyzes user reactions in real time and suggests adjustments to shift schedules.

[1861] Step 7:

[1862] Server: Automatically distributes confirmed shift schedules to employees.

[1863] The server receives the shift schedule that the user has confirmed and modified and automatically distributes it to all employees.

[1864] Processing demand forecasts for services and products

[1865] Step 1:

[1866] Server: Collects customer purchase history and sales data.

[1867] The server collects customer purchase history and sales data from the database every day at 11:00 p.m.

[1868] Step 2:

[1869] Server: Preprocesses the collected data and imputes missing or outlier values.

[1870] After data collection, the server performs preprocessing, imputing missing values ​​and removing outliers.

[1871] Step 3:

[1872] Server: Inputs the pre-processed data into the AI ​​model to predict future demand.

[1873] The server passes the pre-processed data to an AI model to predict future demand.

[1874] Step 4:

[1875] Server: Based on the prediction results, a list of recommended promotion targets and inventory replenishment is generated.

[1876] The server automatically generates a list of recommended promotional targets and inventory replenishment based on the prediction results output by the AI ​​model.

[1877] Step 5:

[1878] User: Review the generated recommendation list and incorporate it into promotion planning and inventory replenishment.

[1879] The user checks the recommended list and implements promotion plans and inventory replenishment as necessary.

[1880] Step 6:

[1881] Emotion engine: Analyzes user reactions and suggests adjustments to the recommendation list.

[1882] The emotion engine analyzes user reactions in real time and makes suggestions to adjust the recommendation list.

[1883] In this way, the invention will realize sustainable service provision through food waste reduction, labor shortage mitigation, and appropriate pricing. In addition, by integrating an emotion engine, more accurate prediction and optimization will be possible, leading to improved user satisfaction.

[1884] Example 2

[1885] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1886] In conventional business operation systems, customer purchasing information, inventory data, disposal data, and employee shift information are managed separately, making data integration and management cumbersome and making efficient operation difficult. Furthermore, predictions and optimization based on this data are often performed manually, limiting the efficiency of operations. Furthermore, because feedback that takes user emotions into account is not reflected, optimization results may not meet on-site needs. Therefore, a system that can solve these issues all at once is needed.

[1887] The identification processing by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting customer purchasing information, inventory data, disposal data, and employee shift information, means for preprocessing the collected data and complementing missing values ​​and outliers, means for performing demand forecasting using a generative AI model based on the preprocessed data, means for optimizing order quantities, optimizing pricing, and automatically generating employee shifts based on the generated demand forecast, means for obtaining user feedback using a sentiment analysis engine based on the prediction results and the generated data and adjusting necessary information, and means for automatically executing necessary actions based on information confirmed and corrected by the user. This enables integrated data management and efficient business operations, and realizes highly accurate predictions and optimization that reflect user feedback.

[1888] "Customer purchasing information" refers to data such as the products purchased by the customer, the purchase date and time, and the quantity purchased.

[1889] "Inventory data" refers to data including information such as product inventory quantity, date of receipt, and storage location.

[1890] "Disposal data" refers to data that includes information such as the quantity of inventory items to be discarded, the reason for disposal, and the date and time of disposal.

[1891] "Employee shift information" refers to data such as employees' working days, working hours, and desired holidays.

[1892] "Means of collection" refers to the methods and equipment used to obtain the necessary information from databases and sensors.

[1893] "Preprocessing means" refers to a method for completing missing data values, eliminating outliers, and converting the data into a format suitable for analysis.

[1894] A "generative AI model" refers to a model that includes algorithms for demand forecasting and optimization from data using techniques such as machine learning and deep learning.

[1895] "Means for demand forecasting" refers to methods and systems that use AI models to estimate future demand.

[1896] "Order quantity optimization" refers to the process of determining the optimal product order quantity based on demand forecasts.

[1897] "Pricing optimization" refers to the process of determining the most effective price by taking into account the balance between supply and demand.

[1898] "Automatic employee shift generation" refers to a system that automatically creates work schedules based on employee preferences and business needs.

[1899] An "emotion analysis engine" refers to a technology that analyzes emotions from user i...

Claims

1. A means for collecting customer purchasing information, inventory data, disposal data, and employee shift information; A means of preprocessing the collected data and imputing missing and outlier values; A means for using AI models based on the pre-processed data to forecast demand, optimize inventory orders, optimize pricing, and automatically generate employee shifts; and A means for presenting necessary information to a user based on the prediction results and generated data, and for the user to confirm and correct the information; A system that includes a means to automatically execute necessary actions based on information confirmed and corrected by the user.

2. A means to use an AI model to forecast demand for the next week based on the pre-processed data and calculate order quantities; A means for automatically generating an order list based on the calculated order quantity; 2. The system of claim 1, further comprising means for automatically transmitting the order list confirmed and modified by the user to the supplier.

3. Based on the pre-processed data, an AI model is used to analyze the relationship between demand and price, and to set the price for the next day.

2. The system according to claim 1, further comprising means for automatically transmitting and reflecting the set price in a POS system.

4. Using the pre-processed data, an AI model is used to predict optimal shifts based on the number of employees and peak times, and a means is used to automatically generate shift schedules.

2. The system according to claim 1, further comprising means for automatically distributing the shift schedule confirmed and modified by the user to employees.

5. Using pre-processed data, an AI model is used to predict future demand and generate a list of recommended promotions and inventory replenishment.

2. The system according to claim 1, further comprising means for automatically executing a promotion plan and inventory replenishment based on the recommendation list confirmed and modified by the user.

Images